Linguistic findings in persons with schizophrenia—a review of the current literature

Abstract

Introduction:

Alterations of verbalized thought occur frequently in psychotic disorders. We characterize linguistic findings in individuals with schizophrenia based on the current literature, including findings relevant for differential and early diagnosis.

Methods:

Review of literature published via PubMed search between January 2010 and May 2022.

Results:

A total of 143 articles were included. In persons with schizophrenia, language-related alterations can occur at all linguistic levels. Differentiating from findings in persons with affective disorders, typical symptoms in those with schizophrenia mainly include so-called “poverty of speech,” reduced word and sentence production, impaired processing of complex syntax, pragmatic language deficits as well as reduced semantic verbal fluency. At the at-risk state, “poverty of content,” pragmatic difficulties and reduced verbal fluency could be of predictive value.

Discussion:

The current results support multilevel alterations of the language system in persons with schizophrenia. Creative expressions of psychotic experiences are frequently found but are not in the focus of this review. Clinical examinations of linguistic alterations can support differential diagnostics and early detection. Computational methods (Natural Language Processing) may improve the precision of corresponding diagnostics. The relations between language-related and other symptoms can improve diagnostics.

1 Introduction

Language comprehension and production are among the most complex brain functions in humans and seem to follow fundamental governing principles (Sportiche et al., 2014). By communicating, language provides a link between the externally perceivable and internal thought processes. In psychiatry, observations of altered language are traditionally taken as indications of underlying alterations in thought. In particular, schizophrenia can be accompanied by severe alterations of language functions. Idiosyncratic psychotic experiences can also lead to unusual verbal expressions that are often highly creative (Heinz, 2023). However, traditional accounts understand linguistic alterations in persons with schizophrenia as deficits directly caused by the psychotic disorder, and our review will also focus on this approach, while recognizing the need to also address the creative use of language in future studies.

With respect to a deficit-oriented account of linguistic alterations, explanatory models of verbal expression as a manifestation of an impaired thought process go back to, among others, the theoretical accounts of Kraepelin (1896) and Bleuler (1911a). According to Bleuler, the coincidence of disturbed regular associations and random combinations of only superficially related ideas would lead to “disjointed” (“zerfahren”), “bizarre,” “incorrect,” and “abrupt thinking” with “loss of goal” (Bleuler, 1911b). Further, an (especially subjective) increase of associatively altered thoughts (“Gedankendrängen”) was contrasted to an (objective and subjective) “blocking of thoughts” (“Sperrung”; Kraepelin, 1896), which—together with “poverty of ideas”—Bleuler regarded as fundamental for the symptomatology (Bleuler, 1911c). In addition to these association disorders, he classified, e.g., a “wrong” choice of words, a disturbed formation of new and “wrong” words as well as grammar violations as genuine linguistic disorders, which in the most severe cases could lead to completely incomprehensible “word salad” (Bleuler, 1911c). These symptoms—which could also be found in a mild form in healthy people—could occur only episodically in people with schizophrenia, sometimes with mild symptoms, sometimes with clear disturbances, up to high-grade association disturbances, which would ultimately result in incoherence (“Zerfahrenheit”) (Bleuler, 1911b). Jaspers (1946) distinguished incoherence from an often witty or creative flight of ideas. Yet, seemingly incoherent or only loosely connected utterances are found in modern poetry or may result from anxiety and differences in social status, power and education (Wübben, 2012; Heinz, 2023). Accordingly, Schneider (1950) and Muleh and Carpenter (1974) and in this tradition ICD-10 (World Health Organization, 1993) focuses on language-related symptoms reported by the patients themselves (such as thought insertion or complex acoustic hallucinations) rather than impressions of coherence by psychiatrists—a critical approach that will be addressed in the discussion of this review.

A systematization of language-related findings as “positive” (i.e., pressure of speech, tangentiality, derailment, incoherence and illogicality) and “negative” (i.e., poverty of speech and poverty of content of speech) symptoms using the Scale for the Assessment of Thought, Language, and Communication (TLC) (Andreasen, 1979a; see Box 1) reflects evolutionary speculations of (Jackson 1884) (for a criticism see Heinz, 1998). It promotes a characterization of subtypes and severity levels across diagnostic groups (Andreasen, 1979b; Andreasen and Grove, 1986).

Conceptually, clinically describable formal thought disorders (FTD) can be understood as a “multidimensional construct” encompassing alterations in thinking (including organization, control, processing, and expression), language, and communication (Hart and Lewine, 2017) (for an illustration see Figure 1).

Figure 1

The language-related manifestation of FTD motivated descriptions on the basis of linguistic categories. These generally comprise phonology, morphology, syntax, lexical semantics, and pragmatics with their respective sub-processes (e.g., Ahlsén, 2006). From this perspective, individuals with schizophrenia have been found to exhibit primarily altered pragmatics, with, e.g., impaired cohesion and coherence, as well as altered lexical access, formation of neologisms, stilted speech, and association chaining (Covington et al., 2005). In contrast, morphology and basic syntax have been described as largely unaffected (Covington et al., 2005), as has phonology (Covington et al., 2005, cf.Parola et al., 2023, but cf.de Boer et al., 2023; Voppel et al., 2023).

Reflecting the conceptual levels of thought production and verbalized expression, “disorganized speech” (American Psychiatric Association, 2013) and “thought disorder” (WHO, 2019) entered the current diagnostic manuals as leading symptoms of schizophrenia and are recorded as positive and negative FTD in common clinical scales—especially in the Scale for the Assessment of Negative Symptoms (SANS) (Andreasen, 1983) and the Scale for the Assessment of Positive Symptoms (SAPS) (Andreasen, 1984) as well as the Positive and Negative Syndrome Scale (PANSS) (Kay et al., 1987) and the Structured Interview for Psychosis-risk Syndromes (SIPS) (McGlashan et al., 2001). In addition, a number of language-specific scales have been developed to assess FTD in detail (for a selection see Box 1).

1.1 Aim of the review study and search strategy

While we acknowledge and respect the creativity in linguistic expressions of psychotic experiences, this review focuses on linguistic findings considered as pathological in persons with schizophrenia to approach the following questions:

• Which linguistic phenomena are found in persons with schizophrenia?

• Which findings have predictive utility?

• Is it possible to distinguish more disease-specific from transdiagnostic phenomena?

For this purpose, recent investigations of FTD on the basis of clinical scales as well as investigations on the linguistic levels of lexico-semantics, syntax, and pragmatics will be presented. In addition, an overview of the currently rapidly growing findings from natural language processing (NLP) will be given followed by studies focusing on early detection and comparative studies.

A literature search was performed via PubMed [PubMed (nih.gov)] with limitation of the search period from 2010 to May 2022 using the following search terms:

• (language) OR (speech) AND (schizophrenia); this yielded 2,724 results

• (speech) OR (language) AND (differential diagnosis) AND (schizophrenia) (all fields); this yielded 71 results

• (speech) OR (language) AND (differential diagnosis) AND (psychosis) (all fields); this yielded 68 results

• (language) AND (differential diagnosis) AND (psychiatry) (all fields); this yielded 216 results; of these, those related to schizophrenia/psychotic disorders were manually selected

• (“formal thought disorder”) AND (“differential diagnosis”) (all fields): this yielded 2 results

• Inclusion criteria were:

• Studies focusing on clinical speech or language assessment (including production or processing) in adults diagnosed with schizophrenia, psychotic disorders, or schizophrenia spectrum disorder

• Studies focusing on clinical speech or language assessment in persons at high risk of schizophrenia/psychosis

• Studies focusing on speech or language assessment in adults diagnosed with schizophrenia compared with persons diagnosed with either bipolar disorder and/or schizoaffective disorders, and/or depressive disorder

• Primary Research

• Reviews and Meta-Analyses that reflected the most current research were included

• Article written in English or German

• Exclusion Criteria were:

• Case Reports

• Studies focusing on children/adolescents diagnosed with schizophrenia/psychosis

• Studies with main focus on neuroimaging or genetics

• Studies exclusively on motor speech function/prosody

• Studies exclusively on “inner speech” or acoustic hallucinations

• Studies exclusively on speech therapy

• Studies focusing on the specifics of a particular language (including sign language) or on bilingualism

• Studies on body language

• Studies of texts of affected persons published on the internet (e.g., on “social media”)

• After deleting duplicates, this resulted in a total of 143 included studies (including 11 reviews and 5 meta-analyses; all articles were written in English).

• Regarding terminology, the term “psychosis” has been adopted in this review when used by the authors of the reported studies without more specificity.

2 Current clinical findings

2.1 Findings based on clinical scales

Studies from the current literature search reported prevalences of 55% FTD in the early stages of psychotic disorders (Oeztuerk et al., 2022) or 19% FTD at first episode of schizophrenia (Roche et al., 2015), and overall 27% clinically impairing FTD in schizophrenia (Roche et al., 2015). Conceptually, the presence of FTD already in early phases of psychotic illness has been suggested to be a “marker of ‘psychosis proneness’,” whereas the acute manifestation of FTD may represent a “breakdown” in language structure (Roche et al., 2015).

The discriminability between positive FTD (especially tangentiality, derailment, incoherence, pressure of speech, and illogical thinking) and negative FTD was replicated in current studies using both language-specific scales (TLC in Comparelli et al., 2020, TALD in Kircher et al., 2014) and the SAPS/SANS (Roche et al., 2016; Ucok et al., 2021). This repeatedly showed a predominance of positive FTD in the early phases of psychotic illness, which tend to decline over time, whereas negative FTD often persist in the further course of the disease (Roche et al., 2016; Comparelli et al., 2020; Ucok et al., 2021).

Whereas the prevalence of positive FTD has been associated with worse clinical outcome [in terms of overall symptomatology (Wilcox et al., 2012; Little et al., 2019), hospitalization (Roche et al., 2016), and occupational functioning (Muralidharan et al., 2018)], contradictory results suggested both independence from cognitive functions (Comparelli et al., 2020; Fuentes-Claramonte et al., 2021) versus associations with specific cognitive impairments [involving goal maintenance (Becker et al., 2012), executive functions (Little et al., 2019; Oeztuerk et al., 2022), problem solving (Little et al., 2019), and attention (Oeztuerk et al., 2022)]. Clinically, positive FTD tended to intensify when talking about negatively connoted content (so-called affective reactivity of language; Docherty et al., 1994; Minor et al., 2016).

Negative FTD, on the other hand, has been associated not only with decreased quality of life (Bowie et al., 2011; Tan et al., 2014), but also more unequivocally with cognitive impairments [involving verbal working memory (Becker et al., 2012; Nagels et al., 2016; Ucok et al., 2021), problem-solving (Comparelli et al., 2020), and attention (Nagels et al., 2016; Ucok et al., 2021)] and illogicality (as assessed by the TLC) (Comparelli et al., 2020). Moreover, the global severity of FTD seemed related to impairments in social functioning (Comparelli et al., 2020; Oeztuerk et al., 2021; meta-analysis, see Marggraf et al., 2020), where negative FTD could determine the observable behavior and positive FTD the performance (“appropriateness”) in social interactions (as assessed by the Specific Level of Function Scale; Schneider and Struening, 1983; Bowie et al., 2011). Lastly, “peculiar use of words” and “peculiar logic” in the Thought and Language Index (TLI) (Liddle et al., 2002, see Box 1) that contributed to the distinction between individuals with and without first psychotic episode (Ayer et al., 2016) or with and without clinical remission (Yalınçetin et al., 2016) may have independent diagnostic value.

In summary assessments by clinical scales indicate a preponderance of positive FTD in the early phase of schizophrenia, while negative FTD tend to prevail in the further course, leading to impairments in about a third of the patients. While studies found both to be associated with further disease-related impairment in cognitive and social functioning, positive FTD have among others been related to greater clinical burden and negative FTD to lower quality of life.

2.2 Findings from linguistic analyses

2.2.1 Lexico-semantics

Analyses at the lexico-semantic level usually assume a network model of semantic memory, in which the strength of connections between individual semantic “entries” (memory items) is determined by their shared use or conceptual similarity (Collins and Loftus, 1975; Levelt, 1999; Rofes et al., 2019). According to this, word retrieval at the initial level of abstract conceptualization (e.g., Binder and Desai, 2011; Kiefer and Pulvermüller, 2012) and the subsequent level of lexico-semantic access (e.g., Pulvermüller, 1999) leads to automatic spreading activation within the network, so that co-activated associated entries will be accessed more quickly (so-called priming) and must in turn be inhibited during the final selection (Collins and Loftus, 1975; Dell, 1986).

2.2.1.1 Lexico-semantics (processing)

Against this background, lexical decision tasks, which require participants to indicate whether a target stimulus presented after a so-called prime word is a real word or not (e.g., house vs. frouse), allow quantification of word relatedness effects. Here the presentation of target words related to the prime leads to shorter reaction times than the presentation of unrelated words (Meyer and Schvaneveldt, 1971). This is expressed by the priming effect (Neely, 1991), where hyperpriming refers to an excessive difference between reaction time for unrelated compared to related words and hypopriming to a reduced priming effect.

In simultaneous EEG recordings of event-related potentials (ERPs), an additional reduction in the amplitude of the N400 component is observed when related words are presented, the so-called N400 effect (Bentin et al., 1985). In this context, the N400—a negativity with peak amplitude at approximately 400 ms after stimulus onset—most likely reflects the neuronal activity resulting from a temporary “binding” between the stimulus and the pre-activated semantic network (Kutas and Federmeier, 2011).

Early influential findings indicated hyperpriming in persons with schizophrenia with FTD (as assessed by “conceptual disorganization” in the Brief Psychiatric Rating Scale;Overall and Gorham, 1962) both when semantic stimuli were directly and only indirectly [by an implicit mutual association, e.g., “lemon-(sour)-sweet”] related (Spitzer, 1993; Spitzer et al., 1993). The latter was evoked specifically at short interstimulus intervals (the so-called stimulus onset asynchrony; SOA), which are thought to provoke automatic semantic activation processes rather than controlled activation (which is provoked by longer intervals). This was suggested as “indicator of associative network dysfunction” (Spitzer et al., 1993) with an “overinclusive” network leading not only to a more readily activation of close but also of more distant associations (Spitzer, 1993).

Regarding more recent findings, a current review reported heterogeneous results (Almeida and Radanovic, 2021): Hyperpriming, described by the majority of studies and associated with positive FTD (e.g., Safadi et al., 2013; Kuperberg et al., 2019), was repeatedly reported (Almeida and Radanovic, 2021), whereas other studies found persistent (i.e., beyond the acute phase of illness; Besche-Richard et al., 2014) hypopriming (e.g., Niznikiewicz et al., 2010; Sass et al., 2014; Tan et al., 2015). Furthermore, the typicality effect (i.e., faster response to typical than to atypical words) was found to be reduced during the initial episode of schizophrenia spectrum disorder (but not during stabilization) and correlated with negative symptoms (Hui et al., 2012).

Also with regard to N400 findings, review papers pointed to incongruent study results (Kuperberg et al., 2010; Mohammad and DeLisi, 2013) without a clear correlation between specific N400 patterns and FTD measures (Almeida and Radanovic, 2021). Here, persons with schizophrenia predominantly showed a reduced N400 effect (e.g., Laurent et al., 2010; Mathalon et al., 2010; Niznikiewicz et al., 2010; Kiang et al., 2014; meta-analysis see Wang et al., 2011; review see Almeida and Radanovic, 2021), which has been described as “normalized” after the acute phase of disease (Besche-Richard et al., 2014). A modulation by the SOA seems to be important here. Thus, a reduced N400 effect tended to be detectable with long SOAs, which, in contrast to short intervals, favor controlled search processes (i.e., using strategies such as prediction; Kuperberg et al., 2010), whereas short SOAs tended to lead to statistically normal or enhanced N400 effects (Wang et al., 2011; Mohammad and DeLisi, 2013). Also the presentation of homonyms in their sub-dominant meaning (e.g., bank: riverbank = sub-dominant; financial institution = dominant) led to a diminished N400 effect, which has been interpreted as a preferential preselection or a reduced suppression of dominant word meanings in persons with schizophrenia (Salisbury, 2010). This finding could reflect a reduced impact of semantic context, which can suppress dominant word meanings in specific circumstances (Heinz et al., 2019). Lastly, also the presentation of high-frequency words (i.e., words that occur frequently in a language) led to markedly large N400 amplitudes, indicating a disruption of their preferential activation (Condray et al., 2010).

2.2.1.2 Lexico-semantics (production)

Verbal fluency (VF) tests require the production of as many correct words as possible in a short period of time. Thus, an underlying alternation between automatic spreading activation within the semantic network and active retrieval strategies using working memory and executive functions is assumed (Troyer et al., 1997; Turner, 1999; review see Henry and Crawford, 2005). Whereas semantic VF (i.e., the production of words belonging to a given semantic category) is thought to be based more on intact semantic memory functions and in particular on activation of language-relevant regions in the temporal area, phonemic VF (i.e., the production of words starting with a given letter) seems to depend more on frontal executive functions (Henry and Crawford, 2004; Baldo et al., 2006). In addition to the total number of words, the number and size of so-called clusters (i.e., the rapid production of usually related words) and switches (i.e., slow changes between mostly unrelated words) are commonly assessed (Gruenewald and Lockhead, 1980; Troyer et al., 1997; Ehlen et al., 2020).

In persons with schizophrenia, deficits in semantic VF have been repeatedly shown with high effect sizes, whereas phonemic VF was little or not affected (Neill et al., 2014; Tan et al., 2015; meta-analysis see Knowles et al., 2010; Schaefer et al., 2013). As depicted in Figure 2, both semantic and phonemic VF deficits have been associated with working memory dysfunction (Ojeda et al., 2010). Phonemic VF deficits have additionally been related to negative symptoms (Ojeda et al., 2010). Deficits in semantic VF, on the other hand, have been associated with both positive FTD (Egeland et al., 2018) and—mediated by cognitive speed (as measured by the Digit Symbol Substitution Test; Wechsler, 1997)—alogia (Brébion et al., 2018) as well as with disease duration (Ojeda et al., 2010). Semantic tasks furthermore revealed above-average production of early acquired words with high typicality (expressed as percentage values based on normed data from categorization tasks) (Juhasz et al., 2012). Apart from that, in semantic tasks, persons with schizophrenia produced comparatively few and small clusters of semantically related concepts (Berberian et al., 2016; Tan et al., 2020), which differed in their semantic cluster patterns from those typically observed in healthy controls (Sung et al., 2012). Here, a lower number of related words was associated with positive FTD, whereas negative FTD were associated with increased response latencies (Docherty et al., 2011) as well as low word and switch counts (Egeland et al., 2018).

Figure 2

These results have been interpreted to support the presumed relationship between disorganized language and semantic memory structure disorders on the one hand, and negative FTD and executive dysfunction on the other (Docherty et al., 2011). Hypothetically, a negative correlation between switches in phonematic VF tasks and cerebellar GABA levels (quantified by magnetic resonance spectroscopy) in persons with schizophrenia may reflect a possible overinhibition within the prefronto-cerebellar system, which might normally contribute to the suppression of competing alternatives in VF tasks (Piras et al., 2019). In addition to the classic semantic and phonematic tasks, VF deficits also affected verb production (Badcock et al., 2011; Smirnova et al., 2017). Over a one-year course of illness, VF deficits appeared unchanged (Grimes et al., 2021) and seemed to persist across studies independently of other disease-related factors in persons with first psychotic episode, chronic schizophrenia, as well as their first-degree relatives (Tan et al., 2020).

Furthermore, pronounced deficits in picture naming tests for verbs and nouns as a function of picture complexity have been described (Kambanaros et al., 2010). Finally, categorical word production tasks yielded inconsistent results with evidence for the production of less typical responses by persons with schizophrenia (Brébion et al., 2010, but cf.Brébion et al., 2013) that could be related to thought disorganization (Brébion et al., 2013, but cf.Brébion et al., 2010). Affective flattening, on the other hand, has been associated with unimpaired use of typical words and interpreted as a possible inhibition of commission errors (Brébion et al., 2010, 2013). In this context, also neuroleptic effects have to be taken into account that cause secondary negative symptoms, particularly affective flattening and apathy, due to impairment of dopamine neurotransmission (Heinz et al., 1998).

2.2.2 Syntax

Disturbances in the processing of complex syntax have been replicated several times in recent studies in the sense of decreasing sentence comprehension with increasing syntactic complexity (Dwyer et al., 2014; Tan et al., 2016; Çokal et al., 2019; Tan and Rossell, 2019; di San et al., 2022). Corresponding deficits have been associated with impairments in working memory (di San et al., 2022). However, given the suggested relationship between syntactic comprehension deficits and positive FTD (particularly derailment, circumstantiality, and loss of goal; Tan et al., 2016), syntactic processing deficits have been proposed as a subcomponent of positive FTD (Tan and Rossell, 2019; meta-analysis see Bora et al., 2019). It has been reported that such impairments are particularly evident in individuals with schizophrenia with FTD (Çokal et al., 2019), although they also occurred to a lesser extent in persons with schizophrenia without FTD, which may support a presumed continuum of context-dependent processing deficits in schizophrenia (Dwyer et al., 2014).

The current literature search identified no studies evaluating syntax production in individuals with schizophrenia, except for those comparing schizophrenia with bipolar disorder. The corresponding results are summarized in 2.5.1.

2.2.3 Pragmatics

Pragmatics is traditionally distinguished from semantics and syntax by describing the use and comprehension of linguistic expressions in an interactional context that enables communication using common reference points including, e.g., abstract terms, idioms, and allusions (Morris, 1938). Up to 96% of individuals with schizophrenia have been found to exhibit deficits in subdomains of pragmatics (Bambini et al., 2016), while generally all domains of linguistic, extra-and paralinguistic, contextual, and communicative aspects of both comprehension and production may be affected (Colle et al., 2013). Corresponding deficits were associated with (socio)cognitive dysfunctioning (Bambini et al., 2016) and overall symptom severity (Colle et al., 2013). Logical reasoning, on the other hand, has also been suggested to be impaired, but studies are inconsistent (Gottesman and Chapman, 1960; Mujica-Parodi et al., 2000).

2.2.3.1 Pragmatics (processing)

The process of contextual comprehension is conceivable as the ongoing computation of semantic relations between individual words and their matching against semantic memory with parallel combinatorial processing that ultimately enables the integration of lexico-semantic with syntactic and contextual information (Kuperberg et al., 2010). In persons with schizophrenia, depending on the complexity of the linguistic requirements, there could be an imbalance with a predominance of lexico-semantic matching to the detriment of combinatorial-integrative analysis, which could explain difficulties, e.g., with semantic ambiguities (Kuperberg et al., 2010).

Individuals with schizophrenia have classically been described to have an increased literal proverb understanding, traditionally labeled “concretistic” (Chapman, 1960), which has been replicated using metaphors (Deamer et al., 2019; review see Rossetti et al., 2018), idioms (review see Sela et al., 2015), or irony (review see Rapp et al., 2013). Here, the relevant context that indicates whether an utterance is to be taken literally or reflects a proverb may fail to impact on language processing (Heinz et al., 1996), in accordance with theories that precision of prior knowledge is reduced in psychotic states (Adams et al., 2016; Heinz et al., 2019).

Recent studies indicated low impairment if a choice answer was provided (Ketteler et al., 2012), whereas the formulation of an individual explanation—particularly of proverbs—led to significant impairments (Mercado et al., 2011; Bambini et al., 2020). Metaphor and humor comprehension seemed selectively preserved in individuals with first-episode schizophrenia (Pawełczyk et al., 2018), whereas individuals with a longer history of illness showed impairments in all domains of pragmatics (Pawełczyk et al., 2018). In this context, an association with a disease-related cognitive impairments was shown in persons with first psychotic episode (Perlini et al., 2018). With reference to social functions, proverb comprehension correlated with social cognition (Comparelli et al., 2020), and an association was shown between misinterpretation of metaphors and faulty evaluation of socially appropriate behaviors (Fukuhara et al., 2017).

Regarding the directionality of the disorders, individuals with schizophrenia were suggested to have intact bottom-up processing at the levels of (sub)-lexical sentence and discourse comprehension but impaired top-down processing (Stephane et al., 2014). Conversely, disturbances in the recognition of word relations (antonyms, homonyms, hyperonyms, and synonyms) were already detectable at the lexical level even without sentence context, and correlated with symptom severity (Ketteler et al., 2012).

Also at the sentence and discourse level, the N400 component usually shows larger amplitudes in the presence of contextual incongruence (Kutas and Federmeier, 2011). Individuals with schizophrenia exhibited an early component during which the N400 depended exclusively on contextual congruence and a later component that depended exclusively on semantic associations, whereas healthy individuals showed parallel processing at both levels (Ditman et al., 2011). This was attributed to an impairment of initial activation and later inhibition of semantic associations, whereby individuals with schizophrenia might use the context initially but fail to maintain it when conflicting semantic associations arise (Ditman et al., 2011). Moreover, a reduced N400 effect when lexical associations were not available despite congruent contextual information pointed to impaired discourse comprehension due to excessive reliance on semantic associations (Swaab et al., 2013).

2.2.3.2 Pragmatics (production)

The formation of linguistic references, e.g., between mentioned characters, temporal, causal, and local information plays a crucial role for interference-free communication and corresponding impairments are called disorders of referential coherence (Ditman et al., 2011). Coherence implies a common frame of reference between the interlocutors. In persons with schizophrenia, disorders of referential coherence can manifest in disjointed speech flow with diminished thematic references between words and sentences, and also in the violation of conversational rules like quantity, relevance, regard of shared vs. new information etc. (Grice, 1975; Corcoran and Frith, 1996; Brüne and Bodenstein, 2005; Kuperberg, 2010a,b). Coherence in relation to the overarching content and joint reference frame between sentences or parts of speech, can be distinguished in part from cohesion, which rather refers to the surface structure of a sentence. Cohesion results from the immediate establishment of connections between concepts through the use of appropriate conjunctions, pronouns, substitutions, etc. (Halliday and Hasan, 1976).

In recent studies, individuals with schizophrenia and clinical FTD showed failures in the use of referential pronouns (Sevilla et al., 2018), unclear linguistic references (Çokal et al., 2018), as well as contextually inappropriate word substitutions and misinterpretations in repetition tasks, most likely due to impaired word access or impaired contextual integration (Dwyer et al., 2014). At the communication level, there were difficulties in relating to the current communication topic and pronounced disturbances in language production in informal vs. formal contexts (Colle et al., 2013). Also, there were associations between reduced use of conjunctions and reduced capacity for metacognition (Buck et al., 2015), and between impaired Theory-of-Mind and reduced alignment of word use with the communication partner (Dwyer et al., 2020) (here, alignment means that the speaker and the listener align their linguistic representations, thus facilitating communication). In contrast to this observation, unimpaired alignment was described in another study (Sharpe et al., 2022), and social interactions appeared to enhance reference formation, semantic cohesion, and contextualization (Saavedra, 2010).

With regard to the ability to infer the either concrete or abstract meaning of an utterance according to context, production might be less impaired than comprehension: For example, the use of abstract expressions in free speech was only slightly reduced in individuals with schizophrenia despite impaired metaphor comprehension (Elvevåg et al., 2011). Also the completion of given phrases was unimpaired despite a slowing associated with negative symptomatology (Pesciarelli et al., 2014). In contrast, marked deficits in the domain of irony have been described in patients with chronic schizophrenia not only in comprehension but also in production (Colle et al., 2013).

In summary, linguistic analyses delineated impairments on the levels of lexico-semantics, syntax and pragmatics. Results on lexico-semantic processing showed hyperpriming likely associated with positive FTD, hypopriming, and mainly reduced N400 effects. Regarding word production, especially deficits in semantic VF point to a disease-related dysfunction of semantic memory, which seemed persistent, could intensify as a function of disease duration and was also observed in first-degree relatives. Corresponding deficits appear to affect mainly the automatic activation of closely related items and may have a greater impact on neural connections formed in the later phase of word acquisition.

Individuals with schizophrenia typically exhibit deficits in processing complex syntax. This has been associated with working memory deficits and positive—but not negative—FTD and may also occur in the absence of FTD.

Concerning pragmatic language comprehension, the long-standing finding was confirmed that explaining inferential meaning (e.g., in idioms or irony) is impaired. This could particularly affect individuals with disease-related cognitive impairments. Corresponding deficits could be closely related to impairments in social cognition, suggesting a common underlying alteration in context-dependent understanding of semantic expressions. This could be attributed to predominant processing based on semantic relations with reduced contextual integration, possibly due to impaired top-down processing. At the level of pragmatic language production, especially impaired use of coherence and referential cohesion was found, which may interfere with social communication.

2.3 Findings from natural language processing

With the aim of improving the objectivity of language-related findings, various NLP approaches have been developed in recent years. In particular, these have been investigated for their utility in characterizing psychosis-typical language features and for their predictive value for the development of psychotic episodes (reviews see Corcoran et al., 2020; Hitczenko et al., 2021). Machine learning has been widely used for this purpose (Corcoran et al., 2018; Perlini et al., 2018; Rezaii et al., 2019; Girard et al., 2021; Sarzynska-Wawer et al., 2021; Voppel et al., 2021; Ziv et al., 2022). In the following, NLP findings related to individuals diagnosed with psychotic disorders are presented first. Findings on early diagnosis and comparative studies are summarized in the subsequent sections.

2.3.1 NLP-findings in persons diagnosed with psychotic disorders

After Latent Semantic Analysis (LSA) (see Box 2) was first used to distinguish individuals with schizophrenia/schizoaffective disorder from healthy persons on the basis of reduced discourse coherence with 80% accuracy (Elvevåg et al., 2007), the method has been widely used and adapted (reviews see Corcoran et al., 2020; Hitczenko et al., 2021). For example, using CoVec (see Box 2) resp. LSA, associations have been demonstrated between reduced semantic coherence in VF tests and derailment, tangentiality (Pauselli et al., 2018), severity of psychotic symptoms, and impairments in psychosocial functioning (Holshausen et al., 2014). Also in terms of reduced coherence, increased variance of word relatedness in free speech has been demonstrated in individuals with schizophrenia spectrum disorder using Word2Vec (see Box 2), allowing discrimination from healthy individuals with approximately 85% accuracy even in the presence of low positive symptomatology (Voppel et al., 2021). In an own investigation, 60% of patients with non-affective psychoses and clinical positive FTD classified correctly by automatically derived coherence metrics; however, the addition of ambiguous referential markers and neologisms improved the model (Just et al., 2023). The newer analysis methods BERT (Devlin et al., 2019) and ELMo (Peters et al., 2018) incorporate more contextual information into their analyses (see Box 2). Using the BERT, decreased question-answer coherence was found, indicating increased tangency, and based on frequent use of incomplete words, assignment to a group with schizophrenia/schizoaffective disorder was made with 90% accuracy, although FTD was clinically detectable in only 20% (Tang et al., 2021). Using the ELMo, texts could be assigned to persons with schizophrenia with over 80% accuracy, especially when many words with low semantic content (e.g., “somehow,” “well”), few positive emotions, or many spiritual words were used (Sarzynska-Wawer et al., 2021). Worth mentioning, in the same study, correct assignment was achieved with over 70% accuracy using the TLC. Moreover, persons with (non-acute) schizophrenia were identified by the production of more words and more errors, with both variables correlating with positive FTD and with positive symptoms in the PANSS without correlation with negative symptoms (Tan et al., 2021). Similarly, exploratory cohesion analyses with Coh-Metrix (McNamara et al., 2014, see Box 2) indicated an association between increased use of pronouns, causal and temporal conjunctions in persons with schizophrenia and linguistic disorganization (Mackinley et al., 2021). On the other hand, part-of-speech tagging was able to assign texts to persons with schizophrenia with 80–90% prediction accuracy due to, e.g., reduced verbs, adverbs, and adjectives, more diverse lexemes, and more frequent use of the first person singular (Ziv et al., 2022). This has been interpreted as less complex, more associative, and more self-centered language and may be indicative of poverty of content and reduced cohesion. Regarding decreased linguistic cohesion (determined by Coh-Metrix) as a marker for negative FTD, a statistical path model implied that impaired causal/intentional cohesion—partly direct, partly mediated by decreased self-reflectivity (as measured by the Metacognition Assessment Scale-Abbreviated; Lysaker et al., 2005)—may lead to simplified language that is difficult for the other person to understand (García-Mieres et al., 2020). In addition, both using Coh-Metrix (Mackinley et al., 2021) and the Analytic Thinking Index (Silva et al., 2021), persons with first-episode schizophrenia were shown to have greater narrativity (i.e., episodic, narrative, and less formal language style), which correlated with disorganized thinking in the PANSS and the TLI and may indicate a lower capacity for hierarchical structuring of thoughts. Finally, emotion-related autobiographical texts have been attributed to individuals with schizophrenia on the basis of lower expressivity and complexity, more self-reference, and more repetition especially when referring to anger and joy (Hong et al., 2015).

In summary, NLP studies particularly showed reduced semantic and discourse coherence in persons with schizophrenia, which seems to be related to positive FTD, but is also detectable in clinically low-grade FTD. In this regard, high discrimination accuracy between individuals with and without psychotic disorders has been repeatedly shown. Also the production of incomplete and content-poor words, low use of verbs, adverbs, and adjectives, reduced expressiveness and complexity, a less formal but more narrative style of language, and increased self-reference seem to be typical of schizophrenia. Finally, the production of more words overall (especially conjunctions and pronouns) and more errors could be specifically related to positive symptoms.

2.4 Early detection and predictive value

The predictive value of linguistic analyses was first demonstrated in adolescents at clinical high risk with a specific association between “illogical” thinking (assessed as “inappropriate” causal or noncausal utterances or statements which are simultaneously made and refuted by the Kiddie Formal Thought Disorder Rating Scale, K-FTDS; Caplan et al., 1989), but see for inconsistent results (Gottesman and Chapman, 1960; Mujica-Parodi et al., 2000), poverty of content (as assessed by the K-FTDS), as well as decreased referential cohesion (as assessed by a modified analysis of cohesion; Halliday and Hasan, 1976) and the manifestation of psychosis (Bearden et al., 2011). Comparably, an association between disorganization and manifestation of psychosis has been repeatedly shown in adults at clinical high risk using clinical scales (e.g., Demjaha et al., 2012; DeVylder et al., 2014; review see Oeztuerk et al., 2022). Studies at the levels of lexico-semantics, syntax, and pragmatics revealed findings in individuals at high risk that were comparable to those of persons with schizophrenia, and in some cases, associated with an increased risk for manifestation of psychosis. Comparable to persons with schizophrenia (Wang et al., 2011; Mohammad and DeLisi, 2013), in persons at clinical high risk, the N400 effect in priming studies was largely intact at short SOAs but markedly diminished at long intervals (Lepock et al., 2019), which was associated with a decrease in global functioning levels (Lepock et al., 2022). This was interpreted in terms of intact early activation of related concepts but poor maintenance of such activation and suggested as a possible neurophysiological biomarker for schizophrenia risk (Lepock et al., 2019). Individuals at clinical high-risk furthermore showed marked impairment in semantic but not phonematic VF (Becker et al., 2010; Magaud et al., 2010), predictive of transition to manifest psychosis (Becker et al., 2010). On the other hand, an association between decreased phonematic VF and transition to schizophrenia has also been demonstrated (Pawełczyk et al., 2021). Underlying the VF deficits may be a specific impairment in word selection despite intact single-word retrieval, which in individuals at ultra-high-risk correlated with both disorganized and negative symptoms (Vargas et al., 2018).

More equivocal findings emerged from comprehensive studies of pragmatics: Here, individuals at ultra-high-risk showed significant impairments in understanding inferential meanings, in discourse analysis, and in the socio-emotional domain, but better performance in metaphor explanation and in the cognitive domain than individuals with first psychotic episode (Pawełczyk et al., 2019). However, transition to schizophrenia was specifically associated with impairments in humor comprehension and metaphor explanation (Pawełczyk et al., 2021).

Also NLP-studies indicated some similarities to the above findings in individuals with schizophrenia: Thus, correlations were reported between low semantic coherence and positive FTD (tangentiality, derailment, and circumstantiality) as well as between reduced lexico-semantic complexity and negative FTD in individuals with clinical high risk using the BERT (Bilgrami et al., 2022).

Using word2vec, individuals at clinical high risk showed reductions in coherence and on-topic scores (as a measure of tangentiality) comparable to those of individuals with first psychotic episode (Morgan et al., 2021). That said, this pattern was most apparent in picture descriptions (rather than during free speech) and was no longer significant when controlling for IQ and schooling (Morgan et al., 2021). Beyond that, reduced cohesion in the form of reduced word stem overlap was detectable by Coh-Metrix in individuals at ultra-high-risk, which was associated with an increase in positive, negative, and disorganized symptoms in the SIPS as well as impairments in verbal learning (Gupta et al., 2018). Furthermore, individuals at genetic high risk for schizophrenia showed increased atypical lexical associations, which correlated with higher schizotypy scores, but occurred in only one of five picture description tasks (Manschreck et al., 2012). In addition, building on a proof-of-principle study (Bedi et al., 2015) transition from a prodromal phase to a psychotic episode was predicted with 80% accuracy by LSA and part-of-speech tagging, with primarily lower semantic coherence and lower use of possessive pronouns identified as critical distinguishing features (Corcoran et al., 2018). Here, the computational linguistic findings correlated with illogical thinking, content poverty (both clinically assessed with the K-FTDS), and low referential cohesion (Corcoran et al., 2018). In contrast, a joint analysis of LSA-based semantic coherence, part-of-speech tagging, brain structural findings, and functional connectivity did not allow discrimination between individuals at clinical high-risk and healthy individuals (Haas et al., 2020). Similarly, during treatment of an acute psychotic episode (in schizophrenia, bipolar disorder, major depression, or a related disorder), no predictive value of NLP-based lexical, sentence coherence, or disfluency measures could be demonstrated in relation to positive or negative symptoms (Girard et al., 2021). Yet, correlation were shown between positive symptoms and increased use of perceptual process words indicative of perceptual distortion, as well as between negative symptoms and decreased use of relative words suggesting poverty of speech, increased negative emotion words, and error corrections (Girard et al., 2021).

As a further NLP approach, speech graphs were investigated with regard to their predictive value for psychotic disorders. In this method, which is based on graph theory, a network-like modeling of individual linguistic utterances is carried out independently of semantic or syntactic features (Mota et al., 2012; see Box 3). The performance of the method has been evaluated as exceptionally high, but with uncertainty as to which anomalies it reflects and how valid it is (Hitczenko et al., 2021). Compared with individuals at clinical high risk, the connectivity of speech graphs from picture descriptions was reduced (indicated by lower LCC, LCCr, LSCr; see Box 3) in individuals with a first psychotic episode and correlated with negative FTD in the TLI (Morgan et al., 2021; Spencer et al., 2021). Notably, LSC and LSCr were also significantly reduced in those who developed psychosis within the following 7 years, suggestive of a relationship between speech connectedness and transition to psychosis (Morgan et al., 2021; Spencer et al., 2021). Finally, the NLP methods Vector Unpacking and Latent Context Analysis (see Box 3) were able to achieve a predictive ability of 90% for the later manifestation of schizophrenia in persons in the prodromal phase. These findings were based on low semantic density (as a measure of content poverty), which correlated with negative symptoms, and increased references to voices, whispers, and sounds that correlated with positive symptoms (Rezaii et al., 2019). Since the vector-based computation was strongly correlated with the assessment of semantic density by human raters, this method was also found to be particularly promising (Hitczenko et al., 2021).

In summary, primarily disorganization, some degree of illogical thinking, poverty of content, and reduced referential cohesion as assessed by clinical scales as well as impaired semantic and phonemic VF, and difficulties in understanding humor and explaining metaphors may be associated with conversion from the at-risk state to first psychotic episode. NLP studies furthermore pointed to reduced semantic coherence (possibly best detectable with the BERT), reduced connectivity of speech graphs, reduced semantic density, and increased references to auditory sensations as possible predictors of conversion to psychosis.

2.5 Comparative studies

2.5.1 Schizophrenia compared to bipolar disorder

Already Bleuler pointed out the necessity of distinguishing between disjointed speech in persons with schizophrenia and severe forms of flight of ideas with accelerated thinking in persons in manic episodes (Bleuler, 1911b). Andreasen and Grove showed a preponderance of (prognostically unfavorable) negative FTD in individuals with schizophrenia compared with tendentially stronger and transient positive FTD in individuals with mania (Andreasen and Grove, 1986).

Recent comparative studies that used clinical scales have supported the hypothesis of more frequent negative FTD in persons with schizophrenia and also showed persistent positive FTD to occur more frequently in them than in persons with bipolar disorder (e.g., Wilcox et al., 2012; meta-analysis see Yalincetin et al., 2017). However, dimension reduction procedures argued for the conceptualization of FTD as a continuum with a highly specific distribution of the factors “Verbosity,” “Poverty of Speech,” and “Disorganization” with specifically increased verbosity and overall more frequent FTD in first-time manic episodes vs. first manifestation of schizophrenia (45% vs. 19%) (Roche et al., 2015, cf.Cuesta and Peralta, 2011). When groups were matched for overall levels of positive FTD and cognitive performance, differences consisted solely of more poverty of content of speech and perseveration in the group with schizophrenia (Kircher et al., 2022).

At the lexico-semantic level, individuals with schizophrenia made more errors than those with bipolar disorder when assessing word associations (e.g., “honey” and “stings” both counting as associated with “bee”), with erroneous associations between unrelated terms being related to FTD severity and missing associations being related to negative symptoms (Jamadar et al., 2013). When explicitly asked to judge word relatedness, preserved N400 effects were measured in individuals with schizophrenia (Ryu et al., 2012; Wang et al., 2020), with decreased N400 amplitudes overall (Wang et al., 2020) or specifically for unrelated words (Ryu et al., 2012). In contrast, conflicting results indicated comparatively large N400 amplitudes for related words with reduced N400 effect (Ryu et al., 2012) or overall slightly reduced N400 amplitudes with preserved N400 effect (Wang et al., 2020) in persons with bipolar disorder. Moreover, a preceding presentation of homophones elicited a magnification of the N400 congruency effect in people with schizophrenia, whereas it was diminished in healthy individuals and those with bipolar disorder (Raucher-Chéné et al., 2019).

Compared to individuals with bipolar disorder, word production in semantic (but not phonemic, Meesters et al., 2013) VF tasks was reduced in individuals with schizophrenia (Esan et al., 2020)—especially in the acute phase, but also in remission (Ceylan et al., 2020) and in elderly (Meesters et al., 2013). The former group did not differ from healthy controls (Ceylan et al., 2020) or performed worse (Meesters et al., 2013; Esan et al., 2020).

At the syntactic level, both individuals with schizophrenia and bipolar disorder made more comprehension errors than healthy controls (Perlini et al., 2012). Here, individuals with schizophrenia produced shorter sentences, made more paragrammatic errors, and exhibited lower syntactic completeness (Perlini et al., 2012). The amplitude of the ERP component P600, which is associated with a second-parse attempt to understand syntactically invalid sentences (cf.Kuperberg et al., 2010), was reduced both in individuals with schizophrenia and bipolar disorder, which was associated with a reduction in the P600 effect, pointing to a failure of contextual integration in both disorders (Lee et al., 2016).

At the discourse level, coherence was decreased in persons with schizophrenia compared to healthy individuals, especially in the case of severe overall symptoms, whereas no differences were found compared to individuals with bipolar disorder (who themselves did not differ significantly from healthy individuals) (Perlini et al., 2012).

Speech graphs in particular have been used as NLP methods to differentiate between individuals with schizophrenia and bipolar disorder. In a first pilot study, when applied to dream reports, they showed a discriminability between both groups with more than 90% accuracy (Mota et al., 2012). In particular, decreased connectivity was described in persons with schizophrenia (Mota et al., 2012; Palaniyappan et al., 2019), as well as increased knots and edges and frequent changes in contextual references in persons with mania (Mota et al., 2012). Exclusively in persons with schizophrenia, associations were shown between low LSCr (i.e., lower referential ties, see Box 3) and global functional level, and between low LCC and LCCr (i.e., lower speech connectedness, see Box 3) and decentralization of the core brain hubs in fMRI (Palaniyappan et al., 2019). This was interpreted as a biological marker of cerebral dysconnectivity that could condition speech dysconnectivity (Palaniyappan et al., 2019). However, the authors found no correlations with clinical measures of FTD. It was furthermore possible to distinguish between individuals who developed schizophrenia or bipolar disorder 6 months after initial manifestation of a psychotic syndrome with very high accuracy on the basis of a “disorganization index” derived from LCC and LSC, which correlated with negative symptoms in the PANSS and was more pronounced in the antecedent of schizophrenia (Mota et al., 2017). Importantly, the results were obtained exclusively from dream reports and negative picture stories, but not from the history interview or from memory descriptions. In contrast to the above results (Meesters et al., 2013; Esan et al., 2020), application of speech graph analysis to VF tasks showed no significant differences between the two clinical groups, although individuals with schizophrenia presented the expected differences from healthy individuals in terms of lower nodes and edges (Chrobak et al., 2022).

2.5.2 Schizophrenia compared to schizoaffective disorder

No relevant differences were found between individuals with schizophrenia and schizoaffective disorder using the TLC (Wilcox et al., 2012) or comprehensive linguistic assessments (Pantano et al., 2016) and in many of the studies mentioned above, mixed cohorts were summarized (i.e., Bowie et al., 2011; Docherty et al., 2011; Mercado et al., 2011; Becker et al., 2012; Juhasz et al., 2012; Wilcox et al., 2012; Stephane et al., 2014; Tan et al., 2014, 2015, 2016; Smirnova et al., 2017; Muralidharan et al., 2018; Tan and Rossell, 2019; García-Mieres et al., 2020; Girard et al., 2021; Grimes et al., 2021; Tang et al., 2021; Voppel et al., 2021; Chrobak et al., 2022; Sharpe et al., 2022). Nonetheless, a study of communication failure found stronger deficits in persons with (non-acutely psychotic) schizophrenia than with schizoaffective disorder (Docherty et al., 2013). Here, more than 50% of the communication dysfunctions, especially in the group with schizophrenia, were explained by combined as well as single deficits in the areas of neurocognition, emotional perception, and Theory of Mind (Docherty et al., 2013).

2.5.3 Schizophrenia compared to depressive disorder

Individuals with schizophrenia did not differ from (non-age and education matched) individuals with depressive disorders in basic semantic-lexical language skills and showed better comprehension of emotional prosody (Pawełczyk et al., 2020). Yet, in pragmatics, they showed significantly poorer abilities in inferring implicit meanings, understanding humor and metaphors, and discourse comprehension, regardless of symptom severity (Pawełczyk et al., 2020). Beyond that, exclusively in individuals with schizophrenia, a pronounced affective reactivity of language was demonstrable in the form of reduced information reference in negative versus positive valence of topic (Rubino et al., 2011) and in the form of a general increase in FTD when speaking about social relations (Cohen et al., 2014). Fittingly, in persons with schizophrenia, negative but not positive or neutral sentence endings resulted in larger N400 amplitudes than in persons with depression/dysthymia, who in turn did not differ from healthy individuals (Klumpp et al., 2010).

2.5.4 Comparison schizophrenia, bipolar disorder, schizoaffective disorder, affective disorder

Comparative studies that collectively examined FTD in cohorts with schizophrenia, manic and depressive episodes in affective and schizoaffective disorders using the TLC (Cuesta and Peralta, 2011), TALD (Kircher et al., 2014), or SAPS/SANS (Stein et al., 2022) confirmed a transdiagnostic occurrence of FTD with highly specific distribution of distinguishable symptom domains. Specifically, symptoms attributed to the factors „Objective positive “(e.g., derailment, crosstalk, dissociation of thinking, tangentiality, and logorrhea) (Kircher et al., 2014) or “Disorganization” (e.g., derailment, loss of goal, tangentiality, circumstantiality, and illogicality) and “Verbosity” (e.g., clanging, pressure of speech, echolalia, incoherence, and neologisms) (Cuesta and Peralta, 2011) were most pronounced in individuals with manic episodes. In contrast, symptoms attributed to the factors “Negative FTD” (e.g., poverty of thought, inhibited thinking, dysfunction of thought initiative and intentionality, poverty of speech, and slowed thinking) (Kircher et al., 2014) or “Poverty of Speech” (e.g., poverty of content of speech, poverty of speech, and perseveration) (Cuesta and Peralta, 2011) were particularly pronounced in persons with schizophrenia (Cuesta and Peralta, 2011). In addition, frequent “Idiosyncratic Speech” (e.g., word approximations, stilted speech, and neologisms) and “Disorganization” have been described here (Cuesta and Peralta, 2011). Persons with depressive episodes showed both highest scores in “Subjective negative FTD” (e.g., poverty of thought, inhibited thinking, dysfunction of thought initiative and intentionality) (Kircher et al., 2014) and lowest in “Disorganization” and “Verbosity” (Cuesta and Peralta, 2011).

Three factors were identified transdiagnostically with a differential correlation between (1) “Disorganization” and volume reduction in the left temporooccipital language region (comprising parts of the gyrus angularis and the middle occipital gyri) as well as the fractional anisotropy (FA) in the right cingulate and inferior longitudinal fasciculus, between (2) “Emptiness” and volume reduction in the hippocampus and thalamus, and between (3) “Incoherence” and FA in the right cingulate/hippocampal area and reduced FA of the anterior thalamic radiation (Stein et al., 2022).

In summary, the current body of research provides evidence for a transdiagnostic occurrence of clinically detectable FTD in persons with schizophrenia, bipolar, schizoaffective, or affective disorder. Individuals with manic episodes seem most prone to FTD with disease-nonspecific positive FTD (especially verbosity). High discriminability between individuals with schizophrenia and manic episodes was reached on the basis of disease-specific poverty of speech in schizophrenia with reduced word production, shorter sentences, and lower syntactic completeness. Moreover, persons with schizophrenia exhibited idiosyncratic speech and (nonspecific) disorganization. In addition, high discriminatory and predictive power could be achievable on the basis of reduced connectivity of speech graphs in individuals with schizophrenia. Strong overlap of FTD symptoms was reported between persons with schizoaffective disorder and schizophrenia, although the latter may present with more communication disturbances. Compared to individuals with depressive disorder, persons with schizophrenia may especially show impairments in pragmatics and an intensification of FTD typical of schizophrenia during socio-emotional stress.

3 Discussion

The present review focuses on FTD and investigations at the linguistic levels of lexico-semantics, syntax, and pragmatics in persons with schizophrenia. The reviewed literature confirms that individuals with schizophrenia exhibit numerous linguistic alterations, detectable by clinical scales as well as by more specialized linguistic analyses. We will critically discuss these findings with respect to their theoretical and clinical implications.

At the lexico-semantic level, several priming studies in persons with schizophrenia reported hyperpriming of semantically related items (Safadi et al., 2013; Kuperberg et al., 2019; Almeida and Radanovic, 2021), which has been associated with positive FTD (e.g., Safadi et al., 2013; Kuperberg et al., 2019). An underlying excessive activation of semantic neighbors (Brown and Kuperberg, 2015) and—with respect to neurobiological correlates—a dysfunction of inhibitory interneurons (Almeida and Radanovic, 2021) have been suggested. However, in view of the also frequently reported hypopriming (Niznikiewicz et al., 2010; Besche-Richard et al., 2014; Sass et al., 2014; Tan et al., 2015), the overall results appear inconclusive: Hypopriming has been interpreted in terms of executive dysfunction or semantic memory disorganization leading to impaired contextual integration (Tan et al., 2015; cf.Kuperberg, 2010a,b; Almeida and Radanovic, 2021). In this context, results from N400-studies hint at a potentially important effect of the experimental setting, in that reduced N400 effects both in persons diagnosed with schizophrenia (Wang et al., 2011; Mohammad and DeLisi, 2013) and in those at high risk (Lepock et al., 2019, 2022) were observed particularly when the interstimulus interval (SOA) was long. Since these long SOAs favor the use of controlled semantic search strategies (Kuperberg et al., 2010), automatic activation could be overactive (cf.Spitzer, 1993), while controlled access to semantic memory may be disrupted, hypothetically reflected by low N400 effects. On the other hand, a reduction in N400 effect was also observed in one study at short SOA and was associated with more severe positive symptoms (Besche-Richard et al., 2014). Regarding the comparison of persons with schizophrenia versus bipolar disorder, findings of decreased N400 amplitudes (Ryu et al., 2012; Wang et al., 2020) in persons with schizophrenia may further support a specific reduction in the differentiation between semantically related and unrelated items in the sense of a lowered signal-to-noise ratio among persons with schizophrenia. Interestingly, one study observed comparably large N400 amplitudes upon the presentation of high frequency words (Condray et al., 2010), which usually evoke a low N400 amplitude in control subjects (the so-called frequency-effect). The latter is thought to reflect an enhanced processing of words in a given context based on their familiarity (Kutas and Federmeier, 2011). Taken together, the above results could indicate, that under all these circumstances, context-related information may fail to prime specific semantic memory, or that memory structure itself may be altered. The hypothesis of reduced effects of context in psychosis is supported by the observation of a preferential preselection of homonyms in their dominant meaning (Salisbury, 2010), which can be explained as an understanding of the dominant word meaning to the expensive of the context-dependent inferential meaning (Heinz et al., 1996). Therefore, future priming and N400 studies controlling for both SOAs and symptomatology should also assess semantic memory structure as well as specific effects of sematic and spatial context on behavior.

In terms of word production, repeated findings of semantic VF deficits that precede the transition to psychosis (Becker et al., 2010), persist or exacerbate during the course of the illness (Besche-Richard et al., 2014; Tan et al., 2020; Grimes et al., 2021), and occur in first-degree relatives (Tan et al., 2020), suggest a link between altered priming within the semantic memory network or an altered network connectivity and a predisposition to schizophrenia. Indeed, reduced semantic connectivity was indicated by reduced coherence during semantic VF (Pauselli et al., 2018) as well as diminished clusters (Berberian et al., 2016; Tan et al., 2020) with atypical word relatedness (Sung et al., 2012). Importantly, the former was linked to clinically observable derailment and tangentiality and the latter to both positive (Docherty et al., 2011; Egeland et al., 2018) and negative (Brébion et al., 2018) FTD. In this context, the finding of an above-average production of early acquired words (Juhasz et al., 2012) could indicate a specific alteration of semantic network connections formed later during learning history. The preferential production of early acquired items in the general population (the co-called age-of-acquisition effect; for reviews see, e.g., Juhasz, 2005; Elsherif et al., 2023) could result from higher semantic network plasticity during early compared to later word acquisition, leading to more connections, more central positions, and easier access (Elsherif et al., 2023). In healthy individuals, word acquisition during individual development is accompanied by a continuous increase in gray matter density within temporo-parietal regions relevant for semantic and syntactic processes (Richardson et al., 2010). Thus, assuming a disease-associated impairment in the ability to use context (Heinz et al., 1996; Brown and Kuperberg, 2015), impaired context-based learning (Heinz et al., 2019) could contribute to reduced connectivity of late acquired words within the semantic network and their reduced retrieval. It thus seems conceivable that diminished semantic connectivity is an essential prerequisite for less conventionally structured thinking and may at least partially be present even before the first psychotic episode. On a similar note, semantic coherence was also reduced during free speech (Voppel et al., 2021), where it could be a predictor of transition from risk status to psychosis (Corcoran et al., 2018).

Yet, a creative aspect of language production should also be considered: Unusual word associations could be shaped by psychotic experiences that are difficult to verbalize. This could influence association-based responses especially in the case of negative affective content (e.g., related to trauma).

Finally, the effects of cognitive speed (Brébion et al., 2018) and working memory (Ojeda et al., 2010) on VF performance in individuals with schizophrenia need to be taken into account. These are primarily expected to cause prolonged (Docherty et al., 2011) and diminished switches (Egeland et al., 2018), which have been association with negative FTD. Reduced attention (Nagels et al., 2016; Ucok et al., 2021; Oeztuerk et al., 2022) could additionally contribute to reduced VF performance. Similarly, impaired working memory or cognitive load due to the complexity of the items to be retrieved could also have contributed to the above-average production of early acquired words. This should be investigated in future studies.

With respect to syntax, deficits in processing complex constructions have been unequivocally reported in persons with schizophrenia (Dwyer et al., 2014; Tan et al., 2016; Çokal et al., 2019; Tan and Rossell, 2019; di San et al., 2022). In addition to an association with working memory impairment, such deficits have been specifically related to positive FTD (Tan et al., 2016), but were detectable even in the absence of FTD (Dwyer et al., 2014). Considering that in healthy persons the ability to process complex sentences was associated with the maturation of the arcuate fasciculus, which is fully mature only in adulthood (Skeide et al., 2016), in persons with schizophrenia an association between reduced FA of the arcuate fasciculus and impaired syntax processing would seem plausible. However, no corresponding association has been found (Cavelti et al., 2018; meta-analysis see Cavelti et al., 2018).

In the domain of pragmatics, deficits in understanding context-dependent inferential meaning and discourse comprehension were found in the vast majority of affected individuals (Bambini et al., 2016). This seems consistent with the view that impaired use of higher-order context could lead to overreliance on semantic associations, to the detriment of interpretive understanding and ultimately impairing adaptation to complex social situations (Brown and Kuperberg, 2015, cf.Spitzer, 1993). Impaired metaphor explanation due to overreliance on context-unspecific, dominant word meanings was found to precede the conversion to schizophrenia (Heinz et al., 1996; Pawełczyk et al., 2021), suggesting a fundamental aspect of the disposition to psychotic experiences. Consistent with a reduced use of context based comprehension, pragmatic dysfunctions could be attributed to disturbed top-down processing (Stephane et al., 2014). Importantly, the current studies indicated close connections between inferential meaning and social cognition (Bambini et al., 2016; Fukuhara et al., 2017; Comparelli et al., 2020). The linguistically observable impairment in context use when interpreting figures of speech may represent only a partial phenomenon of a broader impairment of contextualization in psychosis. The hypothesis of reduced use of semantic context may also help to explain significantly higher rates of psychosis among migrants, refugees, and persons with minority status (Brandt et al., 2019; Henssler et al., 2020; Selten et al., 2020; Varchmin et al., 2021). Here, unknown information regarding not only semantic but also pragmatic cultural contexts may increase prediction errors and facilitate stress-associated alterations that contribute to psychotic experiences, particularly when exposed to social exclusion, discrimination and racism (Baker et al., 2021; Lazaridou et al., 2023).

Regarding pragmatic language production, current studies mainly indicate impaired use of referential coherence (Elvevåg et al., 2007; Perlini et al., 2012; Colle et al., 2013; Tang et al., 2021; Voppel et al., 2021) and cohesion (Çokal et al., 2018; Sevilla et al., 2018; García-Mieres et al., 2020; Mackinley et al., 2021). This could be due to a disrupted feedback during language production, which is expected to result in less contextual constraint and loosening of associations when disintegrated associations are incorporated into an utterance (Brown and Kuperberg, 2015; Corcoran et al., 2020). Again, experiences of linguistic prowess, cultural familiarity and stress-associated impairments need to be studied in more detail.

Apart from the language dimension, the cognitive dimension is also expected to contribute to FTD (Hart and Lewine, 2017). In fact, both positive and negative FTD have been associated with impaired attention (Nagels et al., 2016; Ucok et al., 2021; Oeztuerk et al., 2022) and problem-solving ability (Little et al., 2019; Comparelli et al., 2020). This could possibly be due to hallucinations or anxiety and should be investigated in future studies. It should be noted, that in persons with schizophrenia, particularly verbal memory functions may be affected by progressive impairments, possibly due to a reduction in gray matter in the temporal lobe, although other cognitive symptoms have been described as tendentially stable or improving over the life course (Heilbronner et al., 2016). Overall, the present results point to the need to control for cognitive performance and schooling in similar studies (cf.Docherty et al., 2013; Morgan et al., 2021; Kircher et al., 2022).

It seems furthermore important, that both negative and positive FTD have been associated with social sequalae (Bowie et al., 2011; Muralidharan et al., 2018; Comparelli et al., 2020; Marggraf et al., 2020; Oeztuerk et al., 2021). With respect to directionality, it is plausible that FTD cause communication impairments that lead to impairments in social functioning. Alternatively, it also seems conceivable that in the course of the disease, impaired communication, impaired social functioning and social exclusion stress may lead to a further impoverishment of the degrees of freedom of thought in the sense of a vicious circle. This seems to be supported by findings indicating compensation for disturbed top-down processes through language alignment (Sharpe et al., 2022) and improved semantic cohesion (Saavedra, 2010) during social communication. In addition, persons with schizophrenia tend to use particularly few words expressing positive emotions (Sarzynska-Wawer et al., 2021). Under socioemotional stress, they may experience a general intensification of FTD (Cohen et al., 2014), an increase in positive FTD (Minor et al., 2016), or show a decrease in information reference (Rubino et al., 2011). To elucidate whether this might be related to negative experiences, it seems prudent to control for depressive and trauma-related symptoms (cf.Varchmin et al., 2021) in future studies. Social stress exposure and the associated “affective reactivity” (Docherty et al., 1994) may also be related to volume reductions that particularly affect regions relevant to language (i.e., left periinsular region) and emotional processes (i.e., amygdala/medial temporal lobe cluster) (Nickl-Jockschat et al., 2011). Indeed, an association was observed between positive FTD and atrophy in areas involved in emotional processes, including the amygdala (e.g., Spalletta et al., 2010). Thus, assuming a disease (and potentially also antipsychotic-treatment-related; Ho et al., 2011; Fusar-Poli et al., 2013; Heilbronner et al., 2016) vulnerability of the associated brain regions, the language system might be particularly sensitive to emotional stress.

Speech graph measures were used in several studies to quantify speech connectivity. This method showed high discriminatory value (Mota et al., 2012; Palaniyappan et al., 2019, but cf.Chrobak et al., 2022) with reduced speech graph connectivity in individuals with schizophrenia compared to individuals with bipolar disorder. This has been interpreted as an indicator of disorganized speech (Mota et al., 2012; Palaniyappan et al., 2019) and may be related to negative symptoms (Mota et al., 2017; Morgan et al., 2021; Spencer et al., 2021, but cf.Palaniyappan et al., 2019). Reduced speech graph connectivity may also predict transition from clinical high risk to first psychotic episode (Morgan et al., 2021; Spencer et al., 2021) and to schizophrenia (Mota et al., 2017). On a critical note, the applicability seems to depend on the analyzed speech material, so that both the relation to clinically relevant speech samples and the generalizability seem somewhat limited. Also, further analyses of correlation with clinical findings would be desirable. In addition to reduced connectivity, reduced semantic content as measured by the ELMo or Vector Unpacking has been shown to be highly specific for schizophrenia (Sarzynska-Wawer et al., 2021) and to have a high predictive value (Rezaii et al., 2019). This again seems to support the view that negative FTD form a core component of schizophrenia.

Although not the focus of this review, given their overlap with language, neuromotor, affective, cognitive, and psychosocial functions, recent speech-and voice-related findings in individuals with schizophrenia will be briefly discussed: Against the background of meta-analytically demonstrable large heterogeneity of study results and strong effects of task condition (Parola et al., 2020), reduced pitch variability and increased utterance duration have been identified as the only cross-linguistically generalizable acoustic features (while longer pause duration and reduced speech rates were replicated in a subpopulation) (Parola et al., 2023). An association with negative symptoms seems of interest here, since reduced pitch variability has been related to monotone speech and interpreted as flat affect, prolonged pause duration to alogia or affective flattening and increased utterance duration to low energy and high vocal effort (Parola et al., 2023). Especially machine learning approaches reported high accuracy in discriminating between individuals with and without schizophrenia (for a review see Parola et al., 2020). In this context, next to longer pause durations (de Boer et al., 2023), persons with schizophrenia spectrum disorder exhibited a more fragmented (indicated by more voiced segments per second; de Boer et al., 2023; Voppel et al., 2023) and more monotonous (indicated by reduced spectral flux variation; de Boer et al., 2023; resp. lower frequency and volume entropy; Tahir et al., 2019) speech, as well as a more strained voice (indicated by reduced mean spectral slope of voiced and unvoiced regions; de Boer et al., 2023). Importantly, the combination of acoustic and lexico-semantic measures indicated the co-occurrence of fragmented speech and decreased semantic connectedness as indicative of schizophrenia spectrum disorder (Voppel et al., 2023).

The present findings focus largely on the deficit aspects of altered language. Given the findings of an above-average proportion of artistic professions among persons with psychotic disorders (Juda, 1949, but cf.Kyaga et al., 2013) or their relatives (Kyaga et al., 2011, 2013), it seems worthwhile to examine creative aspects of altered language. In particular, an overrepresentation of persons with schizophrenia or bipolar disorder among professional writers could be an indicator of beneficial effects of the loosening of traditional context restraints on linguistic creativity (Kyaga et al., 2013). Further research could illuminate this perspective and possible implications for linguistic creativity in therapeutic contexts.

4 Limitations

The current work has several limitations. Most importantly, the search was performed only in PubMed. Studies that are exclusively accessible via other search engines were therefore not considered. Also, studies examining children/adolescents diagnosed with schizophrenia/psychosis were not included, nor were studies primarily focusing on neuroimaging or genetics. This was done to reduce the large number of relevant studies to those with the strongest relationship to clinical work in adult psychiatry. Similarly, by focusing on clinical findings, the biolinguistic models were not presented in great detail, and no new model was derived from the data. Also, potential interventions were not addressed here, but could be of clinical importance and should therefore be included in future studies. Finally, the presentation of the results led to a deficit-oriented view of the affected individuals. For future studies and reviews, it seems advisable to balance the view by also including strengths and potentials of those affected.

5 Conclusion

Alterations at all levels of the language system appear to be inherent in schizophrenia and are evident both clinically as FDT and in specific linguistic analyses. Corresponding schizophrenia-specific findings reviewed here include poverty of speech with reduced word production (e.g., Cuesta and Peralta, 2011) as well as incomplete (Tang et al., 2021) and content-poor words (Sarzynska-Wawer et al., 2021), short and incomplete sentences (Perlini et al., 2012) with overall reduced expressiveness and complexity (Hong et al., 2015). Furthermore, the processing of complex syntax (e.g., Dwyer et al., 2014) as well as pragmatic language comprehension (e.g., Bambini et al., 2020) and production (e.g., Colle et al., 2013) are typically reduced. A literal understanding of a context-specific inferred understanding of unexpected proverbs and metaphors in explanation tasks can point to a general reduction of the impact of context and point to imprecise priors including alterations in semantic memory (Adams et al., 2016; Heinz et al., 2019).

Notably, already in the at-risk state, linguistic phenomena typical of schizophrenia [including disorganization (Demjaha et al., 2012), poverty of content, reduced referential cohesion (Bearden et al., 2011), difficulties in pragmatic language comprehension (Pawełczyk et al., 2019) and reduced semantic coherence (Bilgrami et al., 2022)] can be detected and appear to be of predictive value for conversion to psychosis. Thus, for clinical work, systematic language evaluation can support early detection, differential diagnosis and assessment of the degree of overall impairment. Here, speech analyses based on natural language processing are gaining importance in order to increase objectivity and precision in differential diagnosis (e.g., Mota et al., 2012) and early detection (e.g., Bilgrami et al., 2022). Of particular interest are schizophrenia-typical reduced speech graph connectivity (e.g., Morgan et al., 2021), reduced semantic density, and increased reference to auditory sensations (Rezaii et al., 2019) that may precede the development of psychosis. So far, these methods have been used in scientific studies, but not in clinical practice, and the relationship of the individual parameters to clinical findings appears to be unclear at present. Future studies should also address how to use specific information regarding linguistic alterations for prevention and therapy and how to use creativity in language use when coping with psychotic experiences.

References

• 1

  AdamsR. A.HuysQ. J. M.RoiserJ. P. (2016). Computational psychiatry: towards a mathematically informed understanding of mental illness. J. Neurol. Neurosurg. Psychiatry87, 53–63. doi: 10.1136/jnnp-2015-310737

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 2

  AhlsénE.Introduction to neurolinguistics. Göteborg: John Benjamins Publishing Company (2006).

  • Google Scholar

• 3

  AlmeidaV. N.RadanovicM. (2021). Semantic priming and neurobiology in schizophrenia: a theoretical review. Neuropsychologia163:108058. doi: 10.1016/j.neuropsychologia.2021.108058

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4

  American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 5th ed.Washington, DC: American Psychiatric Association (2013).

  • Google Scholar

• 5

  AndreasenN. C. (1979a). Thought, language, and communication disorders. I. Clinical assessment, definition of terms, and evaluation of their reliability. Arch. Gen. Psychiatry36, 1315–1321. doi: 10.1001/archpsyc.1979.01780120045006

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 6

  AndreasenN. C. (1979b). Thought, language, and communication disorders: II diagnostic significance. Arch. Gen. Psychiatry36, 1325–1330. doi: 10.1001/archpsyc.1979.01780120055007

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7

  AndreasenN. C.The scale for the assessment of negative symptoms (SANS). The British Journal of psychiatry. Iowa City: University of Iowa (1983).

  • Google Scholar

• 8

  AndreasenNC. The scale for the assessment of positive symptoms (SAPS). Iowa City: University of Iowa (1984).

  • Google Scholar

• 9

  AndreasenN. C. (1986). Scale for the assessment of thought, language, and communication (TLC). Schizophr. Bull.12, 473–482. doi: 10.1093/schbul/12.3.473

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10

  AndreasenN. C.GroveW. M. (1986). Thought, language, and communication in schizophrenia: diagnosis and prognosis. Schizophr. Bull.12, 348–359. doi: 10.1093/schbul/12.3.348

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11

  AyerA.YalınçetinB.AydınlıE.SevilmişŞ.UlaşH.BinbayT.et al. (2016). Formal thought disorder in first-episode psychosis. Compr. Psychiatry70, 209–215. doi: 10.1016/j.comppsych.2016.08.005

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 12

  BadcockJ. C.DragovićM.GarrettC.JablenskyA. (2011). Action (verb) fluency in schizophrenia: getting a grip on odd speech. Schizophr. Res.126, 138–143. doi: 10.1016/j.schres.2010.11.004

  • CrossRef
  • Google Scholar

• 13

  BakerS. J.JacksonM.JongsmaH.SavilleC. W. N. (2021). The ethnic density effect in psychosis: a systematic review and multilevel meta-analysis. Br. J. Psychiatry219, 632–643. doi: 10.1192/bjp.2021.96

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14

  BaldoJ. V.SchwartzS.WilkinsD.DronkersN. F. (2006). Role of frontal versus temporal cortex in verbal fluency as revealed by voxel-based lesion symptom mapping. J. Int. Neuropsychol. Soc.12, 896–900. doi: 10.1017/S1355617706061078

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 15

  BambiniV.ArcaraG.BechiM.BuonocoreM.CavallaroR.BosiaM. (2016). The communicative impairment as a core feature of schizophrenia: frequency of pragmatic deficit, cognitive substrates, and relation with quality of life. Compr. Psychiatry71, 106–120. doi: 10.1016/j.comppsych.2016.08.012

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  BambiniV.ArcaraG.BosinelliF.BuonocoreM.BechiM.CavallaroR.et al. (2020). A leopard cannot change its spots: a novel pragmatic account of concretism in schizophrenia. Neuropsychologia139:107332. doi: 10.1016/j.neuropsychologia.2020.107332

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17

  BeardenC. E.WuK. N.CaplanR.CannonT. D. (2011). Thought disorder and communication deviance as predictors of outcome in youth at clinical high risk for psychosis. J. Am. Acad. Child Adolesc. Psychiatry50, 669–680. doi: 10.1016/j.jaac.2011.03.021

  • CrossRef
  • Google Scholar

• 18

  BeckerT. M.CiceroD. C.CowanN.KernsJ. G. (2012). Cognitive control components and speech symptoms in people with schizophrenia. Psychiatry Res.196, 20–26. doi: 10.1016/j.psychres.2011.10.003

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 19

  BeckerH. E.NiemanD. H.DingemansP. M.van de FliertJ. R.De HaanL.LinszenD. H. (2010). Verbal fluency as a possible predictor for psychosis. Eur. Psychiatry25, 105–110. doi: 10.1016/j.eurpsy.2009.08.003

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20

  BediG.FacundoC. C.Slezak FernándezD.SigmanM.MotaN. B.RibeiroS.et al. (2015). Automated analysis of free speech predicts psychosis onset in high-risk youths. NJP Schizophr.1:15030. doi: 10.1038/npjschz.2015.30

  • CrossRef
  • Google Scholar

• 21

  BentinS.McCarthyG.WoodC. C. (1985). Event-related potentials, lexical decision and semantic priming. Electroencephalogr. Clin. Neurophysiol.60, 343–355. doi: 10.1016/0013-4694(85)90008-2

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 22

  BerberianA. A.MoraesG. V.GadelhaA.BrietzkeE.FonsecaA. O.ScarpatoB. S.et al. (2016). Is semantic verbal fluency impairment explained by executive function deficits in schizophrenia?Rev. Bras. Psiquiatr.38, 121–126. doi: 10.1590/1516-4446-2015-1663

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 23

  Besche-RichardC.IakimovaG.Hardy-BayléM.-C.PasserieuxC. (2014). Behavioral and brain measures (N400) of semantic priming in patients with schizophrenia: test-retest effect in a longitudinal study. Psychiatry Clin. Neurosci.68, 365–373. doi: 10.1111/pcn.12137

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24

  BilgramiZ. R.SaracC.SrivastavaA.HerreraS. N.AzisM.HaasS. S.et al. (2022). Construct validity for computational linguistic metrics in individuals at clinical risk for psychosis: associations with clinical ratings. Schizophr. Res.245, 90–96. doi: 10.1016/j.schres.2022.01.019

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25

  BinderJ. R.DesaiR. H. (2011). The neurobiology of semantic memory. Trends Cogn. Sci.15, 111–121. doi: 10.1016/j.tics.2011.10.001

  • CrossRef
  • Google Scholar

• 26

  BleulerE.Dementia praecox oder Gruppe der Schizophrenien. Leipzig: Franz Deuticke. (1911a). Available at: https://ia800305.us.archive.org/13/items/b21296157/b21296157.pdf

  • Google Scholar

• 27

  BleulerE. (1911b). “Dementia praecox oder Gruppe der Schizophrenien - 1. Kapitel - Die Grundsymptome” in Die einfachen Funktionen. Die Assoziationen (Leipzig: Franz Deuticke), 10–31. Available at: https://ia800305.us.archive.org/13/items/b21296157/b21296157.pdf

  • Google Scholar

• 28

  BleulerE. (1911c). “Dementia praecox oder Gruppe der Schizophrenien - 2. Kapitel – die Akzessorischen Symptome” in Die Alteration der Sprache und Schrift (Leipzig: Franz Deuticke), 121–132. Available at: https://ia800305.us.archive.org/13/items/b21296157/b21296157.pdf%0A

  • Google Scholar

• 29

  BoraE.YalincetinB.AkdedeB. B.AlptekinK. (2019). Neurocognitive and linguistic correlates of positive and negative formal thought disorder: a meta-analysis. Schizophr. Res.209, 2–11. doi: 10.1016/j.schres.2019.05.025

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 30

  BowieC. R.GuptaM.HolshausenK. (2011). Disconnected and underproductive speech in schizophrenia: unique relationships across multiple indicators of social functioning. Schizophr. Res.131, 152–156. doi: 10.1016/j.schres.2011.04.014

  • CrossRef
  • Google Scholar

• 31

  BrandtL.HensslerJ.MüllerM.WallS.GabelD.HeinzA. (2019). Risk of psychosis among refugees: a systematic review and meta-analysis. JAMA Psychiatry76, 1133–1140. doi: 10.1001/jamapsychiatry.2019.1937

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 32

  BrébionG.BressanR. A.OhlsenR. I.PilowskyL. S.DavidA. S. (2010). Production of atypical category exemplars in patients with schizophrenia. J. Int. Neuropsychol. Soc.16, 822–828. doi: 10.1017/S1355617710000664

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 33

  BrébionG.Stephan-OttoC.Huerta-RamosE.UsallJ.OchoaS.RocaM.et al. (2013). Abnormal functioning of the semantic network in schizophrenia patients with thought disorganization. An exemplar production task. Psychiatry Res.205, 1–6. doi: 10.1016/j.psychres.2012.08.007

  • CrossRef
  • Google Scholar

• 34

  BrébionG.Stephan-OttoC.OchoaS.NietoL.ContelM.UsallJ. (2018). Verbal fluency in male and female schizophrenia patients: different patterns of association with processing speed, working memory span, and clinical symptoms. Neuropsychology32, 65–76. doi: 10.1037/neu0000394

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35

  BrownM.KuperbergG. R. (2015). A hierarchical generative framework of language processing: linking language perception, interpretation, and production abnormalities in schizophrenia. Front. Hum. Neurosci.9:643. doi: 10.3389/fnhum.2015.00643

  • CrossRef
  • Google Scholar

• 36

  BrüneM.BodensteinL. (2005). Proverb comprehension reconsidered -- ‘theory of mind’ and the pragmatic use of language in schizophrenia. Schizophr. Res.75, 233–239. doi: 10.1016/j.schres.2004.11.006

  • CrossRef
  • Google Scholar

• 37

  BuckB.MinorK. S.LysakerP. H. (2015). Differential lexical correlates of social cognition and metacognition in schizophrenia; a study of spontaneously-generated life narratives. Compr. Psychiatry58, 138–145. doi: 10.1016/j.comppsych.2014.12.015

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 38

  CaplanR.GuthrieD.FishB.TanguayP.David-LandoG. (1989). The kiddie formal thought disorder rating scale: clinical assessment, reliability, and validity. J. Am. Acad. Child Adolesc. Psychiatry28, 408–416. doi: 10.1097/00004583-198905000-00018

  • CrossRef
  • Google Scholar

• 39

  CaveltiM.KircherT.NagelsA.StrikW.HomanP. (2018). Is formal thought disorder in schizophrenia related to structural and functional aberrations in the language network? A systematic review of neuroimaging findings. Schizophr. Res.199, 2–16. doi: 10.1016/j.schres.2018.02.051

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 40

  CaveltiM.WinkelbeinerS.FederspielA.WaltherS.StegmayerK.GiezendannerS.et al. (2018). Formal thought disorder is related to aberrations in language-related white matter tracts in patients with schizophrenia. Psychiatry Res. Neuroimaging279, 40–50. doi: 10.1016/j.pscychresns.2018.05.011

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 41

  CeylanD.AkdedeB. B.BoraE.AktenerA. Y.Hıdıroğlu OngunC.TuncaZ.et al. (2020). Neurocognitive functioning during symptomatic states and remission in bipolar disorder and schizophrenia: a comparative study. Psychiatry Res.292:113292. doi: 10.1016/j.psychres.2020.113292

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 42

  ChapmanL. J. (1960). Confusion of figurative and literal usages of words by schizophrenics and brain damaged patients. J. Abnorm. Soc. Psychol.60, 412–416. doi: 10.1037/h0043371

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  ChrobakA. A.TurekA.MachalskaK.Arciszewska-LeszczukA.Starowicz-FilipA.KrupaA. J.et al. (2022). Graph analysis of verbal fluency tests in schizophrenia and bipolar disorder. Brain Sci.12:166. doi: 10.3390/brainsci12020166

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 44

  CohenA. S.AusterT.CallawayD.MacAulayR. K.MinorK. S. (2014). Neurocognitive underpinnings of language disorder: contrasting schizophrenia and mood disorders. J. Exp. Psychopathol.5, 492–502. doi: 10.5127/jep.034213

  • CrossRef
  • Google Scholar

• 45

  ÇokalD.SevillaG.JonesW. S.ZimmererV.DeamerF.DouglasM.et al. (2018). The language profile of formal thought disorder. NPJ Schizophr.4, 1–8. doi: 10.1038/s41537-018-0061-9

  • CrossRef
  • Google Scholar

• 46

  ÇokalD.ZimmererV.VarleyR.WatsonS.HinzenW. (2019). Comprehension of embedded clauses in schizophrenia with and without formal thought disorder. J. Nerv. Ment. Dis.207, 384–392. doi: 10.1097/NMD.0000000000000981

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 47

  ColleL.AngeleriR.VallanaM.SaccoK.BaraB. G.BoscoF. M. (2013). Understanding the communicative impairments in schizophrenia: a preliminary study. J. Commun. Disord.46, 294–308. doi: 10.1016/j.jcomdis.2013.01.003

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  CollinsA. M.LoftusE. F. (1975). A spreading-activation theory of semantic processing. Psychol. Rev. doi: 10.1037/0033-295X.82.6.407

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 49

  ComparelliA.CoriglianoV.ForcinaF.BargagnaP.MontalbaniB.FalconeG.et al. (2020). The complex relationship among formal thought disorders, Neurocognition, and functioning in nonacutely ill schizophrenia patients. J. Nerv. Ment. Dis.208, 48–55. doi: 10.1097/NMD.0000000000001087

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 50

  CondrayR.SiegleG. J.KeshavanM. S.SteinhauerS. R. (2010). Effects of word frequency on semantic memory in schizophrenia: electrophysiological evidence for a deficit in linguistic access. Int J Psychophysiol.75, 141–156. doi: 10.1016/j.ijpsycho.2009.10.010

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 51

  CorcoranC. M.CarrilloF.Fernández-SlezakD.BediG.KlimC.JavittD. C.et al. (2018). Prediction of psychosis across protocols and risk cohorts using automated language analysis. World Psychiatry17, 67–75. doi: 10.1002/wps.20491

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 52

  CorcoranR.FrithC. D. (1996). Conversational conduct and the symptoms of schizophrenia. Cogn. Neuropsychiatry1, 305–318. doi: 10.1080/135468096396460

  • CrossRef
  • Google Scholar

• 53

  CorcoranC. M.MittalV. A.BeardenC. E.GurR.HitczenkoK.BilgramiZ.et al. (2020). Language as a biomarker for psychosis: a natural language processing approach. Schizophr. Res.226, 158–166. doi: 10.1016/j.schres.2020.04.032

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 54

  CovingtonM. A.Covington innovations software [WWW document]. (2016). Available at: http://www.covingtoninnovations.com/software.html

  • Google Scholar

• 55

  CovingtonM. A.HeC.BrownC.NaçiL.McClainJ. T.FjordbakB. S.et al. (2005). Schizophrenia and the structure of language: the linguist’s view. Schizophr. Res.77, 85–98. doi: 10.1016/j.schres.2005.01.016

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 56

  CuestaM. J.PeraltaV. (2011). Testing the hypothesis that formal thought disorders are severe mood disorders. Schizophr. Bull.37, 1136–1146. doi: 10.1093/schbul/sbr092

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 57

  de BoerJ. N.VoppelA. E.BrederooS. G.SchnackH. G.TruongK. P.WijnenF. N. K.et al. (2023). Acoustic speech markers for schizophrenia-spectrum disorders: a diagnostic and symptom-recognition tool. Psychol. Med.53, 1302–1312. doi: 10.1017/S0033291721002804

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 58

  DeamerF.PalmerE.VuongQ. C.FerrierN.FinkelmeyerA.HinzenW.et al. (2019). Non-literal understanding and psychosis: metaphor comprehension in individuals with a diagnosis of schizophrenia. Schizophr. Res. Cogn.18:100159. doi: 10.1016/j.scog.2019.100159

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 59

  DellG. S. (1986). A spreading-activation theory of retrieval in sentence production. Psychol. Rev.93, 283–321.

  • Google Scholar

• 60

  DemjahaA.ValmaggiaL.StahlD.ByrneM.McGuireP. (2012). Disorganization/cognitive and negative symptom dimensions in the at-risk mental state predict subsequent transition to psychosis. Schizophr. Bull.38, 351–359. doi: 10.1093/schbul/sbq088

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 61

  DevlinJ.ChangM.-W.LeeK.ToutanovaK. (2019). “BERT: pre-training of deep bidirectional transformers for language understanding” in Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Cornell University, arXiv:1810.04805. 4171–4186.

  • Google Scholar

• 62

  DeVylderJ. E.MuchombaF. M.GillK. E.Ben-DavidS.WalderD. J.MalaspinaD.et al. (2014). Symptom trajectories and psychosis onset in a clinical high-risk cohort: the relevance of subthreshold thought disorder. Schizophr. Res.159, 278–283. doi: 10.1016/j.schres.2014.08.008

  • CrossRef
  • Google Scholar

• 63

  di SanB.PietroC.BarbieriE.MarelliM.de GirolamoG.LuzzattiC. (2022). Processing argument structure and syntactic complexity in people with schizophrenia spectrum disorders. J. Commun. Disord.96:106182. doi: 10.1016/j.jcomdis.2022.106182

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 64

  DitmanT.GoffD.KuperbergG. R. (2011). Slow and steady: sustained effects of lexico-semantic associations can mediate referential impairments in schizophrenia. Cogn. Affect. Behav. Neurosci.11, 245–258. doi: 10.3758/s13415-011-0020-7

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 65

  DochertyA. R.BerenbaumH.KernsJ. G. (2011). Alogia and formal thought disorder: differential patterns of verbal fluency task performance. J. Psychiatr. Res.45, 1352–1357. doi: 10.1016/j.jpsychires.2011.04.004

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 66

  DochertyN. M.DeRosaM.AndreasenN. C. (1996). Communication disturbances in schizophrenia and mania. Arch. Gen. Psychiatry53, 358–364. doi: 10.1001/archpsyc.1996.01830040094014

  • CrossRef
  • Google Scholar

• 67

  DochertyN. M.EvansI. M.SledgeW. H.SeibylJ. P.KrystalJ. H. (1994). Affective reactivity of language in schizophrenia. J. Nerv. Ment. Dis.182, 98–102. doi: 10.1097/00005053-199402000-00006

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 68

  DochertyN. M.McCleeryA.DivilbissM.SchumannE.MoeA.ShakeelM. K. (2013). Effects of social cognitive impairment on speech disorder in schizophrenia. Schizophr. Bull.39, 608–616. doi: 10.1093/schbul/sbs039

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 69

  DwyerK.DavidA.McCarthyR.McKennaP.PetersE. (2014). Higher-order semantic processing in formal thought disorder in schizophrenia. Psychiatry Res.216, 168–176. doi: 10.1016/j.psychres.2014.02.011

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 70

  DwyerK.DavidA. S.McCarthyR.McKennaP.PetersE. (2020). Linguistic alignment and theory of mind impairments in schizophrenia patients’ dialogic interactions. Psychol. Med.50, 2194–2202. doi: 10.1017/S0033291719002289

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 71

  DwyerK.PetersE.McKennaP.DavidA.McCarthyR. (2014). Verbatim recall in formal thought disorder in schizophrenia: a study of contextual influences. Cogn. Neuropsychiatry19, 337–358. doi: 10.1080/13546805.2013.870069

  • CrossRef
  • Google Scholar

• 72

  EgelandJ.HolmenT. L.Bang-KittilsenG.BigsethT. T.EnghJ. A. (2018). Category fluency in schizophrenia: opposing effects of negative and positive symptoms?Cogn. Neuropsychiatry23, 28–42. doi: 10.1080/13546805.2017.1418306

  • CrossRef
  • Google Scholar

• 73

  EhlenF.RoepkeS.KlostermannF.BaskowI.GeiseP.BelicaC.et al. (2020). Small semantic networks in individuals with autism spectrum disorder without intellectual impairment: a verbal fluency approach. J. Autism. Dev. Disord.50, 3967–3987. doi: 10.1007/s10803-020-04457-9

  • CrossRef
  • Google Scholar

• 74

  ElsherifM. M.PreeceE.CatlingJ. C. (2023). Age-of-acquisition effects: a literature review. J. Exp. Psychol. Learn. Mem. Cogn.49, 812–847. doi: 10.1037/xlm0001215

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 75

  ElvevågB.FoltzP. W.WeinbergerD. R.GoldbergT. E. (2007). Quantifying incoherence in speech: An automated methodology and novel application to schizophrenia. Schizophr. Res.93, 304–316. doi: 10.1016/j.schres.2007.03.001

  • CrossRef
  • Google Scholar

• 76

  ElvevågB.HelsenK.De HertM.SweersK.StormsG. (2011). Metaphor interpretation and use: a window into semantics in schizophrenia. Schizophr. Res.133, 205–211. doi: 10.1016/j.schres.2011.07.009

  • CrossRef
  • Google Scholar

• 77

  EsanO.OladeleO.AdediranK. I.Abiona TO (2020). Neurocognitive impairments (NCI) in bipolar disorder: comparison with schizophrenia and healthy controls. J. Affect. Disord.277, 175–181. doi: 10.1016/j.jad.2020.08.015

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 78

  Fuentes-ClaramonteP.López-AraquistainL.SarróS.Sans-SansaB.Ortiz-GilJ.MaristanyT.et al. (2021). Brain functional correlates of formal thought disorder in schizophrenia: examining the frontal/dysexecutive hypothesis. Psychol. Med.51, 2446–2453. doi: 10.1017/S0033291720001063

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 79

  FukuharaK.OgawaY.TanakaH.NagataY.NishidaS.HagaD.et al. (2017). Impaired interpretation of others’ behavior is associated with difficulties in recognizing pragmatic language in patients with schizophrenia. J. Psycholinguist. Res.46, 1309–1318. doi: 10.1007/s10936-017-9497-8

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 80

  Fusar-PoliP.SmieskovaR.KemptonM. J.HoB. C.AndreasenN. C.BorgwardtS. (2013). Progressive brain changes in schizophrenia related to antipsychotic treatment? A meta-analysis of longitudinal MRI studies. Neurosci. Biobehav. Rev.37, 1680–1691. doi: 10.1016/j.neubiorev.2013.06.001

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 81

  García-MieresH.LundinN. B.MinorK. S.DimaggioG.PopoloR.CheliS.et al. (2020). A cognitive model of diminished expression in schizophrenia: the interface of metacognition, cognitive symptoms and language disturbances. J. Psychiatr. Res.131, 169–176. doi: 10.1016/j.jpsychires.2020.09.008

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 82

  GirardJ. M.VailA. K.LiebenthalE.BrownK.KilciksizC. M.PennantL.et al. (2021). Computational analysis of spoken language in acute psychosis and mania. Schizophr. Res.245, 97–115. doi: 10.31234/osf.io/zc47p

  • CrossRef
  • Google Scholar

• 83

  GottesmanL.ChapmanL. J. (1960). Syllogistic reasoning errors in schizophrenia. J. Consult. Psychol.24, 250–255. doi: 10.1037/h0042189

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 84

  GriceH. P. (1975). “Logic and conversation” in Syntax and semantics, speech acts. eds. ColeP.MorganJ. L., vol. 3 (New York, NY: Academic Press)

  • Google Scholar

• 85

  GrimesK. M. D.FoussiasG.RemingtonG.Kalahani-BargisK.ZakzanisK. K. (2021). Stability of verbal fluency in outpatients with schizophrenia. Psychiatry Res.295:113528. doi: 10.1016/j.psychres.2020.113528

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 86

  GruenewaldP. J.LockheadG. R. (1980). The free recall of category examples. J. Exp. Psychol. Hum. Learn. Mem.6, 225–240. doi: 10.1037/0278-7393.6.3.225

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 87

  GuptaT.HesposS. J.HortonW. S.MittalV. A. (2018). Automated analysis of written narratives reveals abnormalities in referential cohesion in youth at ultra high risk for psychosis. Schizophr. Res.192, 82–88. doi: 10.1016/j.schres.2017.04.025

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 88

  HaasS. S.DoucetG. E.GargS.HerreraS. N.SaracC.BilgramiZ. R.et al. (2020). Linking language features to clinical symptoms and multimodal imaging in individuals at clinical high risk for psychosis. Eur. Psychiatry63:e72. doi: 10.1192/j.eurpsy.2020.73

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 89

  HallidayM. A. K.HasanR.Cohesion in English. 1st ed.London: Pearson Education Limited (1976).

  • Google Scholar

• 90

  HartM.LewineR. R. J. (2017). Rethinking thought disorder. Schizophr. Bull.43, 514–522. doi: 10.1093/schbul/sbx003

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 91

  HeilbronnerU.SamaraM.LeuchtS.FalkaiP.SchulzeT. G. (2016). The longitudinal course of schizophrenia across the lifespan: clinical, cognitive, and neurobiological aspects. Harv. Rev. Psychiatry24, 118–128. doi: 10.1097/HRP.0000000000000092

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 92

  HeinzA. (1998). Colonial perspectives in the construction of the psychotic patient as primitive man. Crit. Anthropol.18, 421–444. doi: 10.1177/0308275X9801800404

  • CrossRef
  • Google Scholar

• 93

  HeinzA.Das kolonialisierte Gehirn und die Wege der Revolte. 1st ed.Berlin: Suhrkamp; (2023).

  • Google Scholar

• 94

  HeinzA.KnableM. B.CoppolaR.GoreyJ. G.JonesD. W.LeeK. S.et al. (1998). Psychomotor slowing, negative symptoms and dopamine receptor availability--an IBZM SPECT study in neuroleptic-treated and drug-free schizophrenic patients. Schizophr. Res.31, 19–26. doi: 10.1016/S0920-9964(98)00003-6

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 95

  HeinzA.LeferinkK.BühmannY.HeinzeM. (1996). Autismus und Konkretismus - widersprüchliche Konzepte schizophrener Denkstörungen?Fund. Psychiat.10, 54–61.

  • Google Scholar

• 96

  HeinzA.MurrayG. K.SchlagenhaufF.SterzerP.GraceA. A.WaltzJ. A. (2019). Towards a unifying cognitive, neurophysiological, and computational neuroscience account of schizophrenia. Schizophr. Bull.45, 1092–1100. doi: 10.1093/schbul/sby154

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 97

  HenryJ. D.CrawfordJ. R. (2004). A meta-analytic review of verbal fluency performance following focal cortical lesions. Neuropsychology18, 284–295. doi: 10.1037/0894-4105.18.2.284

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 98

  HenryJ. D.CrawfordJ. R. (2005). A meta-analytic review of verbal fluency deficits in depression. J. Clin. Exp. Neuropsychol.27, 78–101. doi: 10.1080/138033990513654

  • CrossRef
  • Google Scholar

• 99

  HensslerJ.BrandtL.MüllerM.LiuS.MontagC.SterzerP.et al. (2020). Migration and schizophrenia: meta-analysis and explanatory framework. Eur. Arch. Psychiatry Clin. Neurosci.270, 325–335. doi: 10.1007/s00406-019-01028-7

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 100

  HitczenkoK.MittalV. A.GoldrickM. (2021). Understanding language abnormalities and associated clinical markers in psychosis: the promise of computational methods. Schizophr. Bull.47, 344–362. doi: 10.1093/schbul/sbaa141

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 101

  HoB.-C.AndreasenN. C.ZiebellS.PiersonR.MagnottaV. (2011). Long-term antipsychotic treatment and brain volumes: a longitudinal study of first-episode schizophrenia. Arch. Gen. Psychiatry68, 128–137. doi: 10.1001/archgenpsychiatry.2010.199

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 102

  HolshausenK.HarveyP. D.ElvevågB.FoltzP. W.BowieC. R. (2014). Latent semantic variables are associated with formal thought disorder and adaptive behavior in older inpatients with schizophrenia. Cortex55, 88–96. doi: 10.1016/j.cortex.2013.02.006

  • CrossRef
  • Google Scholar

• 103

  HongK.NenkovaA.MarchM. E.ParkerA. P.VermaR.KohlerC. G. (2015). Lexical use in emotional autobiographical narratives of persons with schizophrenia and healthy controls. Psychiatry Res.225, 40–49. doi: 10.1016/j.psychres.2014.10.002

  • CrossRef
  • Google Scholar

• 104

  HuiC. L.-M.LongeneckerJ.WongG. H.-Y.TangJ. Y.-M.ChangW.-C.ChanS. K.-W.et al. (2012). Longitudinal changes in semantic categorization performance after symptomatic remission from first-episode psychosis: a 3-year follow-up study. Schizophr. Res.137, 118–123. doi: 10.1016/j.schres.2012.02.010

  • CrossRef
  • Google Scholar

• 105

  JacksonJ. H. (1884). The croonian lectures on evolution and dissolution of the nervous system. Lancet123, 649–652.

  • Google Scholar

• 106

  JamadarS.O’NeilK. M.PearlsonG. D.AnsariM.GillA.JagannathanK.et al. (2013). Impairment in semantic retrieval is associated with symptoms in schizophrenia but not bipolar disorder. Biol. Psychiatry73, 555–564. doi: 10.1016/j.biopsych.2012.07.027

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 107

  JaspersK. (1946). “Die objektiven Leistungen des Seelenlebens (Leistungspsychologie)” in Allgemeine psychopathologie (Berlin: Springer Berlin Heidelberg), 162.

  • Google Scholar

• 108

  JudaA. (1949). The relationship between highest mental capacity and psychic abnormalities. Am. J. Psychiatry106, 296–307. doi: 10.1176/ajp.106.4.296

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 109

  JuhaszB. J. (2005). Age-of-acquisition effects in word and picture identification. Psychol. Bull.131, 684–712. doi: 10.1037/0033-2909.131.5.684

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 110

  JuhaszB. J.ChambersD.SheslerL. W.HaberA.KurtzM. M. (2012). Evaluating lexical characteristics of verbal fluency output in schizophrenia. Psychiatry Res.200, 177–183. doi: 10.1016/j.psychres.2012.06.035

  • CrossRef
  • Google Scholar

• 111

  JustS. A.BröckerA.-L.RyazanskayaG.NenchevI.SchneiderM.BermpohlF.et al. (2023). Validation of natural language processing methods capturing semantic incoherence in the speech of patients with non-affective psychosis. Front. Psych.14:1208856. doi: 10.3389/fpsyt.2023.1208856

  • CrossRef
  • Google Scholar

• 112

  KambanarosM.MessinisL.GeorgiouV.PapathanassopoulosP. (2010). Action and object naming in schizophrenia. J. Clin. Exp. Neuropsychol.32, 1083–1094. doi: 10.1080/13803391003733578

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 113

  KayS. R.FiszbeinA.OplerL. A. (1987). The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull.13, 261–276. doi: 10.1093/schbul/13.2.261

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 114

  KettelerD.TheodoridouA.KettelerS.JägerM. (2012). High order linguistic features such as ambiguity processing as relevant diagnostic markers for schizophrenia. Schizophr. Res. Treatment.2012:825050. doi: 10.1155/2012/825050

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 115

  KiangM.ChristensenB. K.ZipurskyR. B. (2014). Event-related brain potential study of semantic priming in unaffected first-degree relatives of schizophrenia patients. Schizophr. Res.153, 78–86. doi: 10.1016/j.schres.2014.01.001

  • CrossRef
  • Google Scholar

• 116

  KieferM.PulvermüllerF. (2012). Conceptual representations in mind and brain: theoretical developments, current evidence and future directions. Cortex48, 805–825. doi: 10.1016/j.cortex.2011.04.006

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 117

  KircherT.KrugA.StratmannM.GhaziS.SchalesC.FrauenheimM.et al. (2014). A rating scale for the assessment of objective and subjective formal thought and language disorder (TALD). Schizophr. Res.160, 216–221. doi: 10.1016/j.schres.2014.10.024

  • CrossRef
  • Google Scholar

• 118

  KircherT.SteinF.NagelsA. (2022). Differences in single positive formal thought disorder symptoms between closely matched acute patients with schizophrenia and mania. Eur. Arch. Psychiatry Clin. Neurosci.272, 395–401. doi: 10.1007/s00406-021-01263-x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 119

  KlumppH.KellerJ.MillerG. A.CasasB. R.BestJ. L.DeldinP. J. (2010). Semantic processing of emotional words in depression and schizophrenia. Int. J. Psychophysiol.75, 211–215. doi: 10.1016/j.ijpsycho.2009.12.004

  • CrossRef
  • Google Scholar

• 120

  KnowlesE. E. M.DavidA. S.ReichenbergA. (2010). Processing speed deficits in schizophrenia: reexamining the evidence. Am. J. Psychiatry167, 828–835. doi: 10.1176/appi.ajp.2010.09070937

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 121

  KraepelinE. (1896). “Die Erscheinungen des Irreseins - Störungen des Gedankenganges” in Psychiatrie - ein Lehrbuch für Studirende [sic] und Aerzte. 5th ed (Leipzig: Johann Ambrosius Barth), 130–142. Available at: https://gallica.bnf.fr/ark:/12148/bpt6k76636h/f23.item

  • Google Scholar

• 122

  KuperbergG. R. (2010a). Language in schizophrenia part 1: an introduction. Lang. Linguist. Compass.4, 576–589. doi: 10.1111/j.1749-818X.2010.00216.x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 123

  KuperbergG. R. (2010b). Language in schizophrenia part 2: what can psycholinguistics bring to the study of schizophrenia… and vice versa?Lang. Linguist. Compass.4, 590–604. doi: 10.1111/j.1749-818X.2010.00217.x

  • CrossRef
  • Google Scholar

• 124

  KuperbergG. R.KreherD. A.DitmanT. (2010). What can event-related potentials tell us about language, and perhaps even thought, in schizophrenia?Int. J. Psychophysiol.75, 66–76. doi: 10.1016/j.ijpsycho.2009.09.005

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 125

  KuperbergG. R.WeberK.Delaney-BuschN.UstineC.StillermanB.HämäläinenM.et al. (2019). Multimodal neuroimaging evidence for looser lexico-semantic networks in schizophrenia: evidence from masked indirect semantic priming. Neuropsychologia124, 337–349. doi: 10.1016/j.neuropsychologia.2018.10.024

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 126

  KutasM.FedermeierK. D. (2011). Thirty years and counting: finding meaning in the N400 component of the event-related brain potential (ERP). Annu. Rev. Psychol.62, 621–647. doi: 10.1146/annurev.psych.093008.131123

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 127

  KyagaS.LandénM.BomanM.HultmanC. M.LångströmN.LichtensteinP. (2013). Mental illness, suicide and creativity: 40-year prospective total population study. J. Psychiatr. Res.47, 83–90. doi: 10.1016/j.jpsychires.2012.09.010

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 128

  KyagaS.LichtensteinP.BomanM.HultmanC.LångströmN.LandénM. (2011). Creativity and mental disorder: family study of 300,000 people with severe mental disorder. Br. J. Psychiatry199, 373–379. doi: 10.1192/bjp.bp.110.085316

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 129

  LandauerT.DumaisS. (1997). A solution to Plato’s problem: the latent semantic analysis theory of acquisition, induction, and representation of knowledge. Psychol. Rev.104, 211–240. doi: 10.1037/0033-295X.104.2.211

  • CrossRef
  • Google Scholar

• 130

  LaurentJ.-P.KostovaM.PasserieuxC. (2010). N400 and P300 modulation as functions of processing level in schizophrenia patients exhibiting formal thought disorder. Int. J. Psychophysiol.75, 177–182. doi: 10.1016/j.ijpsycho.2009.10.001

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 131

  LazaridouF. B.SchubertS. J.RingeisenT.KaminskiJ.HeinzA.KlugeU. (2023). Racism and psychosis: an umbrella review and qualitative analysis of the mental health consequences of racism. Eur. Arch. Psychiatry Clin. Neurosci.273, 1009–1022. doi: 10.1007/s00406-022-01468-8

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 132

  LeeC. W.KimS. H.ShimM.RyuV.HaR. Y.LeeS. J.et al. (2016). P600 alteration of syntactic language processing in patients with bipolar mania: comparison to schizophrenic patients and healthy subjects. J. Affect. Disord.201, 101–111. doi: 10.1016/j.jad.2016.05.008

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 133

  LepockJ. R.MizrahiR.GerritsenC. J.BagbyR. M.MaheandiranM.AhmedS.et al. (2022). N400 event-related brain potential and functional outcome in persons at clinical high risk for psychosis: a longitudinal study. Psychiatry Clin. Neurosci.76, 114–121. doi: 10.1111/pcn.13330

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 134

  LepockJ. R.MizrahiR.KorostilM.MaheandiranM.GerritsenC. J.DrvaricL.et al. (2019). N400 event-related brain potential evidence for semantic priming deficits in persons at clinical high risk for psychosis. Schizophr. Res.204, 434–436. doi: 10.1016/j.schres.2018.08.033

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 135

  LeveltW. J. M. (1999). Models of word production. Trends Cogn. Sci.3, 223–232. doi: 10.1016/S1364-6613(99)01319-4

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 136

  LiddleP. F.NganE. T. C.CaissieS. L.AndersonC. M.BatesA. T.QuestedD. J.et al. (2002). Thought and language index: an instrument for assessing thought and language in schizophrenia. Br. J. Psychiatry181, 326–330. doi: 10.1192/bjp.181.4.326

  • CrossRef
  • Google Scholar

• 137

  LittleB.GallagherP.ZimmererV.VarleyR.DouglasM.SpencerH.et al. (2019). Language in schizophrenia and aphasia: the relationship with non-verbal cognition and thought disorder. Cogn. Neuropsychiatry24, 389–405. doi: 10.1080/13546805.2019.1668758

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 138

  LysakerP. H.CarcioneA.DimaggioG.JohannesenJ. K.NicolòG.ProcacciM.et al. (2005). Metacognition amidst narratives of self and illness in schizophrenia: associations with neurocognition, symptoms, insight and quality of life. Acta Psychiatr. Scand.112, 64–71. doi: 10.1111/j.1600-0447.2005.00514.x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 139

  MackinleyM.ChanJ.KeH.DempsterK.PalaniyappanL. (2021). Linguistic determinants of formal thought disorder in first episode psychosis. Early Interv. Psychiatry15, 344–351. doi: 10.1111/eip.12948

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 140

  MagaudE.KebirO.GutA.WillardD.ChauchotF.OlieJ.-P.et al. (2010). Altered semantic but not phonological verbal fluency in young help-seeking individuals with ultra high risk of psychosis. Schizophr. Res.123, 53–58. doi: 10.1016/j.schres.2010.05.005

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 141

  ManschreckT. C.MerrillA. M.JabbarG.ChunJ.DelisiL. E. (2012). Frequency of normative word associations in the speech of individuals at familial high-risk for schizophrenia. Schizophr. Res.140, 99–103. doi: 10.1016/j.schres.2012.06.034

  • CrossRef
  • Google Scholar

• 142

  MarggrafM. P.LysakerP. H.SalyersM. P.MinorK. S. (2020). The link between formal thought disorder and social functioning in schizophrenia: a meta-analysis. Eur. Psychiatry63:e34. doi: 10.1192/j.eurpsy.2020.30

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 143

  MathalonD. H.RoachB. J.FordJ. M. (2010). Automatic semantic priming abnormalities in schizophrenia. Int. J. Psychophysiol.75, 157–166. doi: 10.1016/j.ijpsycho.2009.12.003

  • CrossRef
  • Google Scholar

• 144

  McGlashanT. H.WalshB. C.WoodsS. W.AddingtonJ. C.CannonT.WalkerE.Structured interview for psychosis-risk syndromes. New Haven, CT: Yale School of Medicine (2001).

  • Google Scholar

• 145

  McNamaraD.GraesserA.McCarthyP.CaiZ. (2014). Automated evaluation of text and discourse with Coh-Metrix. Cambridge UK: Cambridge University Press.

  • Google Scholar

• 146

  MeestersP. D.SchouwsS.StekM.de HaanL.SmitJ.EikelenboomP.et al. (2013). Cognitive impairment in late life schizophrenia and bipolar I disorder. Int. J. Geriatr. Psychiatry28, 82–90. doi: 10.1002/gps.3793

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 147

  MercadoC. L.JohannesenJ. K.BellM. D. (2011). Thought disorder severity in compromised, deteriorated, and preserved intellectual course of schizophrenia. J. Nerv. Ment. Dis.199, 111–116. doi: 10.1097/NMD.0b013e3182083bae

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 148

  MeyerD. E.SchvaneveldtR. W. (1971). Facilitation in recognizing pairs of words: evidence of a dependence between retrieval operations. J. Exp. Psychol.90, 227–234. doi: 10.1037/h0031564

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 149

  MikolovT.ChenK.CorradoG. S.DeanJ. (2013). “Efficient estimation of word representations in vector space” in ICLR. Cornell University. arXiv:1301.3781

  • Google Scholar

• 150

  MinorK. S.MarggrafM. P.DavisB. J.MehdiyounN. F.BreierA. (2016). Affective systems induce formal thought disorder in early-stage psychosis. J. Abnorm. Psychol.125, 537–542. doi: 10.1037/abn0000156

  • CrossRef
  • Google Scholar

• 151

  MohammadO. M.DeLisiL. E. (2013). N400 in schizophrenia patients. Curr. Opin. Psychiatry26, 196–207. doi: 10.1097/YCO.0b013e32835d9e56

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 152

  MorganS. E.DiederenK.VértesP. E.IpS. H. Y.WangB.ThompsonB.et al. (2021). Natural language processing markers in first episode psychosis and people at clinical high-risk. Transl. Psychiatry11:630. doi: 10.1038/s41398-021-01722-y

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 153

  MorrisC. (1938). “Foundations of the theory of signs” in International encyclopedia of unified science, vol. 1 (Chicago, IL: University of Chicago Press) Available at: https://pure.mpg.de/rest/items/item_2364493/component/file_2364492/content

  • Google Scholar

• 154

  MotaN. B.CopelliM.RibeiroS. (2017). Thought disorder measured as random speech structure classifies negative symptoms and schizophrenia diagnosis 6 months in advance. NPJ Schizophr.3, 1–10. doi: 10.1038/s41537-017-0019-3

  • CrossRef
  • Google Scholar

• 155

  MotaN. B.FurtadoR.MaiaP.CopelliM.RibeiroS. (2014). Graph analysis of dream reports is especially informative about psychosis. Sci. Rep.4:3691. doi: 10.1038/srep03691

  • CrossRef
  • Google Scholar

• 156

  MotaN. B.VasconcelosN. A. P.LemosN.PierettiA. C.KinouchiO.CecchiG. A.et al. (2012). Speech graphs provide a quantitative measure of thought disorder in psychosis. PLoS One7:e34928. doi: 10.1371/journal.pone.0034928

  • CrossRef
  • Google Scholar

• 157

  Mujica-ParodiL. R.MalaspinaD.SackeimH. A. (2000). Logical processing, affect, and delusional thought in schizophrenia. Harv. Rev. Psychiatry8, 73–83. doi: 10.1080/hrp_8.2.73

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 158

  MulehS.CarpenterW. T. (1974). Clinical psychopathology: Kurt Schneider. Translated by M. W. Hamilton. Grune & Stratton, New York, 1959, 173 pages. Reviewed by Salvatore Muleh and William T. Carpenter, Jr. Compr. Psychiatry15, 255–256. doi: 10.1016/0010-440X(74)90024-8

  • CrossRef
  • Google Scholar

• 159

  MuralidharanA.FinchA.BowieC. R.HarveyP. D. (2018). Thought, language, and communication deficits and association with everyday functional outcomes among community-dwelling middle-aged and older adults with schizophrenia. Schizophr. Res.196, 29–34. doi: 10.1016/j.schres.2017.07.017

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 160

  NagelsA.FährmannP.StratmannM.GhaziS.SchalesC.FrauenheimM.et al. (2016). Distinct neuropsychological correlates in positive and negative formal thought disorder syndromes: the thought and language disorder scale in endogenous psychoses. Neuropsychobiology73, 139–147. doi: 10.1159/000441657

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 161

  NeelyJ. (1991). “Semantic priming effects in visual word recognition: a selective review of current findings and theories” in Basic progress in reading: visual word recognition. eds. BesnerD.HumphreysG. (Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.), 264–333.

  • Google Scholar

• 162

  NeillE.GurvichC.RossellS. L. (2014). Category fluency in schizophrenia research: is it an executive or semantic measure?Cogn. Neuropsychiatry19, 81–95. doi: 10.1080/13546805.2013.807233

  • CrossRef
  • Google Scholar

• 163

  Nickl-JockschatT.SchneiderF.PagelA. D.LairdA. R.FoxP. T.EickhoffS. B. (2011). Progressive pathology is functionally linked to the domains of language and emotion: meta-analysis of brain structure changes in schizophrenia patients. Eur. Arch. Psychiatry Clin. Neurosci.261, S166–S171. doi: 10.1007/s00406-011-0249-8

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 164

  NiznikiewiczM.MittalM. S.NestorP. G.McCarleyR. W. (2010). Abnormal inhibitory processes in semantic networks in schizophrenia. Int. J. Psychophysiol.75, 133–140. doi: 10.1016/j.ijpsycho.2009.10.006

  • CrossRef
  • Google Scholar

• 165

  OeztuerkO. F.PigoniA.AntonucciL. A.KoutsoulerisN. (2022). Association between formal thought disorders, neurocognition and functioning in the early stages of psychosis: a systematic review of the last half-century studies. Eur. Arch. Psychiatry Clin. Neurosci.272, 381–393. doi: 10.1007/s00406-021-01295-3

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 166

  OeztuerkO. F.PigoniA.WenzelJ.HaasS. S.PopovicD.RuefA.et al. (2021). The clinical relevance of formal thought disorder in the early stages of psychosis: results from the PRONIA study. Eur. Arch. Psychiatry Clin. Neurosci.272, 403–413. doi: 10.1007/s00406-021-01327-y

  • CrossRef
  • Google Scholar

• 167

  OjedaN.SánchezP.PeñaJ.ElizagárateE.YollerA. B.LarumbeJ.et al. (2010). Verbal fluency in schizophrenia: does cognitive performance reflect the same underlying mechanisms in patients and healthy controls?J. Nerv. Ment. Dis.198, 286–291. doi: 10.1097/NMD.0b013e3181d61748

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 168

  OverallJ. E.GorhamD. R. (1962). The brief psychiatric rating scale. Psychol. Rep.10, 799–812. doi: 10.2466/pr0.1962.10.3.799

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 169

  PalaniyappanL.MotaN. B.OowiseS.BalainV.CopelliM.RibeiroS.et al. (2019). Speech structure links the neural and socio-behavioural correlates of psychotic disorders. Prog. Neuropsychopharmacol. Biol. Psychiatry10, 112–120. doi: 10.1016/j.pnpbp.2018.07.007

  • CrossRef
  • Google Scholar

• 170

  PantanoT.Fu-IL.CuratoloE.Martins BertiniC.ElkisH. (2016). Thought and language disorders in very early onset schizophrenia, schizoaffective disorder and bipolar disorder. Arch. Clin. Psychiatry43, 67–73. doi: 10.1590/0101-60830000000087

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 171

  ParolaA.SimonsenA.BlikstedV.FusaroliR. (2020). Voice patterns in schizophrenia: a systematic review and Bayesian meta-analysis. Schizophr. Res.216, 24–40. doi: 10.1016/j.schres.2019.11.031

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 172

  ParolaA.SimonsenA.LinJ. M.ZhouY.WangH.UbukataS.et al. (2023). Voice patterns as markers of schizophrenia: building a cumulative generalizable approach via a cross-linguistic and meta-analysis based investigation. Schizophr. Bull.49, S125–S141. doi: 10.1093/schbul/sbac128

  • CrossRef
  • Google Scholar

• 173

  PauselliL.HalpernB.ClearyS. D.KuB.CovingtonM. A.ComptonM. T. (2018). Computational linguistic analysis applied to a semantic fluency task to measure derailment and tangentiality in schizophrenia. Psychiatry Res.263, 74–79. doi: 10.1016/j.psychres.2018.02.037

  • CrossRef
  • Google Scholar

• 174

  PawełczykA.Kotlicka-AntczakM.ŁojekE.PawełczykT. (2019). Preliminary study of higher-order language and extralinguistic impairments in individuals with high clinical risk of psychosis and first episode of schizophrenia. Early Interv. Psychiatry13, 369–378. doi: 10.1111/eip.12482

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 175

  PawełczykA.Kotlicka-AntczakM.ŁojekE.RuszpelA.PawełczykT. (2018). Schizophrenia patients have higher-order language and extralinguistic impairments. Schizophr. Res.192, 274–280. doi: 10.1016/j.schres.2017.04.030

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 176

  PawełczykA.ŁojekE.ŻurnerN.Gawłowska-SawoszM.PawełczykT. (2018). Higher-order language dysfunctions as a possible neurolinguistic endophenotype for schizophrenia: evidence from patients and their unaffected first degree relatives. Psychiatry Res.267, 63–72. doi: 10.1016/j.psychres.2018.05.070

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 177

  PawełczykA.ŁojekE.ŻurnerN.Kotlicka-AntczakM.PawełczykT. (2021). Higher order language impairments can predict the transition of ultrahigh risk state to psychosis-An empirical study. Early Interv. Psychiatry15, 314–327. doi: 10.1111/eip.12943

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 178

  PawełczykA.ŁojekE.ŻurnerN.PawełczykT. (2020). Pragmatic language dysfunctions in schizophrenia and depression patients – a preliminary study. Psychiatr. Psychol. Klin.20, 3–12. doi: 10.15557/PiPK.2020.0001

  • CrossRef
  • Google Scholar

• 179

  PerliniC.BellaniM.FinosL.LasalviaA.BonettoC.ScoccoP.et al. (2018). Non literal language comprehension in a large sample of first episode psychosis patients in adulthood. Psychiatry Res.260, 78–89. doi: 10.1016/j.psychres.2017.11.032

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 180

  PerliniC.MariniA.GarzittoM.IsolaM.CerrutiS.MarinelliV.et al. (2012). Linguistic production and syntactic comprehension in schizophrenia and bipolar disorder. Acta Psychiatr. Scand.126, 363–376. doi: 10.1111/j.1600-0447.2012.01864.x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 181

  PesciarelliF.GamberoniT.FerlazzoF.Lo RussoL.PedrazziF.MelatiE.et al. (2014). Is the comprehension of idiomatic sentences indeed impaired in paranoid schizophrenia? A window into semantic processing deficits. Front. Hum. Neurosci.8:799. doi: 10.3389/fnhum.2014.00799

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 182

  PetersM. E.NeumannM.IyyerM.GardnerM.ClarkC.LeeK.et al. (2018). “Deep contextualized word representations” in Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). Cornell University. arXiv:1802.05365

  • Google Scholar

• 183

  PirasF.PirasF.BanajN.CiulloV.VecchioD.EddenR. A. E.et al. (2019). Cerebellar GABAergic correlates of cognition-mediated verbal fluency in physiology and schizophrenia. Acta Psychiatr. Scand.139, 582–594. doi: 10.1111/acps.13027

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 184

  PulvermüllerF. (1999). Words in the brain’s language. Behav. Brain Sci.22, 253–279. doi: 10.1017/S0140525X9900182X

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 185

  RappA. M.LangohrK.MutschlerD. E.KlingbergS.WildB.ErbM. (2013). Isn’t it ironic? Neural correlates of irony comprehension in schizophrenia. PLoS One8:e74224. doi: 10.1371/journal.pone.0074224

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 186

  Raucher-ChénéD.TerrienS.GobinP.GierskiF.CailliesS.KaladjianA.et al. (2019). Differential semantic processing in patients with schizophrenia versus bipolar disorder: an N400 study. Acta Neuropsychiatr.31, 337–342. doi: 10.1017/neu.2019.9

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 187

  RezaiiN.WalkerE.WolffP. (2019). A machine learning approach to predicting psychosis using semantic density and latent content analysis. NPJ Schizophr.9, 1–12. doi: 10.1038/s41537-019-0077-9

  • CrossRef
  • Google Scholar

• 188

  RichardsonF. M.ThomasM. S. C.FilippiR.HarthH.PriceC. J. (2010). Contrasting effects of vocabulary knowledge on temporal and parietal brain structure across lifespan. J. Cogn. Neurosci.22, 943–954. doi: 10.1162/jocn.2009.21238

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 189

  RocheE.CreedL.MacMahonD.BrennanD.ClarkeM. (2015). The epidemiology and associated phenomenology of formal thought disorder: a systematic review. Schizophr. Bull.41, 951–962. doi: 10.1093/schbul/sbu129

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 190

  RocheE.LyneJ.O’DonoghueB.SeguradoR.BehanC.RenwickL.et al. (2016). The prognostic value of formal thought disorder following first episode psychosis. Schizophr. Res.178, 29–34. doi: 10.1016/j.schres.2016.09.017

  • CrossRef
  • Google Scholar

• 191

  RocheE.LyneJ. P.O’DonoghueB.SeguradoR.KinsellaA.HanniganA.et al. (2015). The factor structure and clinical utility of formal thought disorder in first episode psychosis. Schizophr. Res.168, 92–98. doi: 10.1016/j.schres.2015.07.049

  • CrossRef
  • Google Scholar

• 192

  RofesA.MandonnetE.de AguiarV.RappB.TsapkiniK.MiceliG. (2019). Language processing from the perspective of electrical stimulation mapping. Cogn. Neuropsychol.36, 117–139. doi: 10.1080/02643294.2018.1485636

  • CrossRef
  • Google Scholar

• 193

  RossettiI.BrambillaP.PapagnoC. (2018). Metaphor comprehension in schizophrenic patients. Front. Psychol.9:670. doi: 10.3389/fpsyg.2018.00670

  • CrossRef
  • Google Scholar

• 194

  RubinoI. A.D’AgostinoL.SarchiolaL.RomeoD.SiracusanoA.DochertyN. M. (2011). Referential failures and affective reactivity of language in schizophrenia and unipolar depression. Schizophr. Bull.37, 554–560. doi: 10.1093/schbul/sbp108

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 195

  RyuV.AnS. K.HaR. Y.KimJ. A.HaK.ChoH.-S. (2012). Differential alteration of automatic semantic processing in treated patients affected by bipolar mania and schizophrenia: an N400 study. Prog. Neuropsychopharmacol. Biol. Psychiatry38, 194–200. doi: 10.1016/j.pnpbp.2012.03.009

  • CrossRef
  • Google Scholar

• 196

  SaavedraJ. (2010). Quantitative criteria of narrative coherence and complexity in persons with paranoid schizophrenia. J. Nerv. Ment. Dis.198, 349–355. doi: 10.1097/NMD.0b013e3181da47f5

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 197

  SafadiZ.Lichtenstein-VidneL.DobrusinM.HenikA. (2013). Investigating thought disorder in schizophrenia: evidence for pathological activation. PLoS One8:e82882. doi: 10.1371/journal.pone.0082882

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 198

  SalisburyD. (2010). N400 to lexical ambiguity and semantic incongruity in schizophrenia. Int. J. Psychophysiol.75, 127–132. doi: 10.1016/j.ijpsycho.2009.10.002

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 199

  Sarzynska-WawerJ.WawerA.PawlakA.SzymanowskaJ.StefaniakI.JarkiewiczM.et al. (2021). Detecting formal thought disorder by deep contextualized word representations. Psychiatry Res.304:114135. doi: 10.1016/j.psychres.2021.114135

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 200

  SassK.HeimS.SachsO.StraubeB.SchneiderF.HabelU.et al. (2014). Neural correlates of semantic associations in patients with schizophrenia. Eur. Arch. Psychiatry Clin. Neurosci.264, 143–154. doi: 10.1007/s00406-013-0425-0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 201

  SchaeferJ.GiangrandeE.WeinbergerD. R.DickinsonD. (2013). The global cognitive impairment in schizophrenia: consistent over decades and around the world. Schizophr. Res.150, 42–50. doi: 10.1016/j.schres.2013.07.009

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 202

  SchneiderK.Klinische psychopathologie. Stuttgart: Thieme (1950).

  • Google Scholar

• 203

  SchneiderL. C.StrueningE. L. (1983). SLOF: a behavioral rating scale for assessing the mentally ill. Soc. Work Res. Abstr.19, 9–21. doi: 10.1093/swra/19.3.9

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 204

  SelaT.LavidorM.MitchellR. L. C. (2015). A possible contributory mechanism for impaired idiom perception in schizophrenia. Psychiatry Res.229, 1–11. doi: 10.1016/j.psychres.2015.07.021

  • CrossRef
  • Google Scholar

• 205

  SeltenJ.-P.van der VenE.TermorshuizenF. (2020). Migration and psychosis: a meta-analysis of incidence studies. Psychol. Med.50, 303–313. doi: 10.1017/S0033291719000035

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 206

  SevillaG.RossellóJ.SalvadorR.SarróS.López-AraquistainL.Pomarol-ClotetE.et al. (2018). Deficits in nominal reference identify thought disordered speech in a narrative production task. PLoS One13:e0201545. doi: 10.1371/journal.pone.0201545

  • CrossRef
  • Google Scholar

• 207

  SharpeV.SchootL.LewandowskiK. E.ÖngürD.TürközerH. B.HasoğluT.et al. (2022). We both say tomato: intact lexical alignment in schizophrenia and bipolar disorder. Schizophr. Res.243, 138–146. doi: 10.1016/j.schres.2022.02.032

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 208

  SilvaA.LimongiR.MacKinleyM.PalaniyappanL. (2021). Small words that matter: linguistic style and conceptual disorganization in untreated first-episode schizophrenia. Schizophr Bull open.2:sgab010. doi: 10.1093/schizbullopen/sgab010

  • CrossRef
  • Google Scholar

• 209

  SkeideM. A.BrauerJ.FriedericiA. D. (2016). Brain functional and structural predictors of language performance. Cereb. Cortex26, 2127–2139. doi: 10.1093/cercor/bhv042

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 210

  SmirnovaD.ClarkM.JablenskyA.BadcockJ. C. (2017). Action (verb) fluency deficits in schizophrenia spectrum disorders: linking language, cognition and interpersonal functioning. Psychiatry Res.257, 203–211. doi: 10.1016/j.psychres.2017.07.044

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 211

  SpallettaG.SpoletiniI.CherubiniA.RubinoI. A.SiracusanoA.PirasF.et al. (2010). Cortico-subcortical underpinnings of narrative processing impairment in schizophrenia. Psychiatry Res.182, 77–80. doi: 10.1016/j.pscychresns.2009.11.001

  • CrossRef
  • Google Scholar

• 212

  SpencerT. J.ThompsonB.OliverD.DiederenK.DemjahaA.WeinsteinS.et al. (2021). Lower speech connectedness linked to incidence of psychosis in people at clinical high risk. Schizophr. Res.228, 493–501. doi: 10.1016/j.schres.2020.09.002

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 213

  SpitzerM. (1993). The psychopathology, neuropsychology, and neurobiology of associative and working memory in schizophrenia. Eur. Arch. Psychiatry Clin. Neurosci.243, 57–70. doi: 10.1007/BF02191566

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 214

  SpitzerM.BraunU.HermleL.MaierS. (1993). Associative semantic network dysfunction in thought-disordered schizophrenic patients: direct evidence from indirect semantic priming. Biol. Psychiatry34, 864–877. doi: 10.1016/0006-3223(93)90054-H

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 215

  SporticheD.KoopmanH.StablerE.An introduction to syntactic analysis and theory. Chichester: Wiley Blackwell (2014).

  • Google Scholar

• 216

  SteinF.BuckenmayerE.BroschK.MellerT.SchmittS.RingwaldK. G.et al. (2022). Dimensions of formal thought disorder and their relation to gray-and white matter brain structure in affective and psychotic disorders. Schizophr. Bull.48, 902–911. doi: 10.1093/schbul/sbac002

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 217

  StephaneM.KuskowskiM.GundelJ. (2014). Abnormal dynamics of language in schizophrenia. Psychiatry Res.216, 320–324. doi: 10.1016/j.psychres.2014.02.027

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 218

  SungK.GordonB.VannorsdallT. D.LedouxK.PickettE. J.PearlsonG. D.et al. (2012). Semantic clustering of category fluency in schizophrenia examined with singular value decomposition. J. Int. Neuropsychol. Soc.18, 565–575. doi: 10.1017/S1355617712000136

  • CrossRef
  • Google Scholar

• 219

  SwaabT. Y.BoudewynM. A.LongD. L.LuckS. J.KringA. M.RaglandJ. D.et al. (2013). Spared and impaired spoken discourse processing in schizophrenia: effects of local and global language context. J. Neurosci.33, 15578–15587. doi: 10.1523/JNEUROSCI.0965-13.2013

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 220

  TahirY.YangZ.ChakrabortyD.ThalmannN.ThalmannD.ManiamY.et al. (2019). Non-verbal speech cues as objective measures for negative symptoms in patients with schizophrenia. PLoS One14:e0214314. doi: 10.1371/journal.pone.0214314

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 221

  TanE. J.MeyerD.NeillE.RossellS. L. (2021). Investigating the diagnostic utility of speech patterns in schizophrenia and their symptom associations. Schizophr. Res.238, 91–98. doi: 10.1016/j.schres.2021.10.003

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 222

  TanE. J.NeillE.RossellS. L. (2015). Assessing the relationship between semantic processing and thought disorder symptoms in schizophrenia. J. Int. Neuropsychol. Soc.21, 629–638. doi: 10.1017/S1355617715000648

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 223

  TanE. J.NeillE.TomlinsonK.RossellS. L. (2020). Semantic memory impairment across the schizophrenia continuum: a meta-analysis of category fluency performance. Schizophr. Bull. Open1, 1–14. doi: 10.1093/schizbullopen/sgaa054

  • CrossRef
  • Google Scholar

• 224

  TanE. J.RossellS. L. (2019). Language comprehension and neurocognition independently and concurrently contribute to formal thought disorder severity in schizophrenia. Schizophr. Res.204, 133–137. doi: 10.1016/j.schres.2018.08.019

  • CrossRef
  • Google Scholar

• 225

  TanE. J.ThomasN.RossellS. L. (2014). Speech disturbances and quality of life in schizophrenia: differential impacts on functioning and life satisfaction. Compr. Psychiatry55, 693–698. doi: 10.1016/j.comppsych.2013.10.016

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 226

  TanE. J.YellandG. W.RossellS. L. (2016). Characterising receptive language processing in schizophrenia using word and sentence tasks. Cogn. Neuropsychiatry21, 14–31. doi: 10.1080/13546805.2015.1121866

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 227

  TangS. X.KrizR.ChoS.ParkS. J.HarowitzJ.GurR. E.et al. (2021). Natural language processing methods are sensitive to sub-clinical linguistic differences in schizophrenia spectrum disorders. NPJ Schizophr.7:25. doi: 10.1038/s41537-021-00154-3

  • CrossRef
  • Google Scholar

• 228

  TroyerA. K.MoscovitchM.WinocurG. (1997). Clustering and switching as two components of verbal fluency: evidence from younger and older healthy adults. Neuropsychology11, 138–146. doi: 10.1037/0894-4105.11.1.138

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 229

  TurnerM. A. (1999). Generating novel ideas: fluency performance in high-functioning and learning disabled individuals with autism. J. Child. Psychol. Psychiatry40, 189–201. doi: 10.1111/1469-7610.00432

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 230

  UcokA.KarakaşB.ŞahinO. Ş. (2021). Formal thought disorder in patients with first-episode schizophrenia: results of a one-year follow-up study. Psychiatry Res.301:113972. doi: 10.1016/j.psychres.2021.113972

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 231

  VarchminL.MontagC.TreuschY.KaminskiJ.HeinzA. (2021). Traumatic events, social adversity and discrimination as risk factors for psychosis - an umbrella review. Front. Psych.12:665957. doi: 10.3389/fpsyt.2021.665957

  • CrossRef
  • Google Scholar

• 232

  VargasT.SnyderH.BanichM.NewberryR.ShankmanS. A.StraussG. P.et al. (2018). Altered selection during language processing in individuals at high risk for psychosis. Schizophr. Res.202, 303–309. doi: 10.1016/j.schres.2018.06.036

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 233

  VoppelA. E.de BoerJ. N.BrederooS. G.SchnackH. G.SommerI. E. C. (2021). Quantified language connectedness in schizophrenia-spectrum disorders. Psychiatry Res.304:114130. doi: 10.1016/j.psychres.2021.114130

  • CrossRef
  • Google Scholar

• 234

  VoppelA. E.de BoerJ. N.BrederooS. G.SchnackH. G.SommerI. E. C. (2023). Semantic and acoustic markers in schizophrenia-Spectrum disorders: a combinatory machine learning approach. Schizophr. Bull.49, S163–S171. doi: 10.1093/schbul/sbac142

  • CrossRef
  • Google Scholar

• 235

  WangK.CheungE. F. C.GongQ.ChanR. C. K. (2011). Semantic processing disturbance in patients with schizophrenia: a meta-analysis of the N400 component. PLoS One6:e25435. doi: 10.1371/journal.pone.0026739

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 236

  WangK.ZhaoY.-L.TanS.-P.ZhangJ.-G.LiD.ChenJ.-X.et al. (2020). Semantic processing event-related potential features in patients with schizophrenia and bipolar disorder. Psych J.9, 247–257. doi: 10.1002/pchj.321

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 237

  WechslerD.Wechsler adult intelligence scale—3rd edition (WAIS-3®). San Antonio, TX: Harcourt Assessment (1997).

  • Google Scholar

• 238

  WHOICD-11 international classification of diseases for mortality and morbidity statistics 11th revision. (2019). Available at: https://icd.who.int/browse11/l-m/en

  • Google Scholar

• 239

  WilcoxJ.WinokurG.TsuangM. (2012). Predictive value of thought disorder in new-onset psychosis. Compr. Psychiatry53, 674–678. doi: 10.1016/j.comppsych.2011.12.002

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 240

  World Health Organization (1993). The ICD-10 classification of mental and behavioural disorders. Geneva: World Health Organization.

  • Google Scholar

• 241

  WübbenY.Verrückte Sprache - Psychiater und Dichter in der Anstalt des 19. Jahrdunderts. Konstanz: Konstanz University Press (2012).

  • Google Scholar

• 242

  YalincetinB.BoraE.BinbayT.UlasH.AkdedeB. B.AlptekinK. (2017). Formal thought disorder in schizophrenia and bipolar disorder: a systematic review and meta-analysis. Schizophr. Res.185, 2–8. doi: 10.1016/j.schres.2016.12.015

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 243

  YalınçetinB.UlaşH.VarL.BinbayT.AkdedeB. B.AlptekinK. (2016). Relation of formal thought disorder to symptomatic remission and social functioning in schizophrenia. Compr. Psychiatry70, 98–104. doi: 10.1016/j.comppsych.2016.07.001

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 244

  ZivI.BaramH.BarK.ZilbersteinV.ItzikowitzS.HarelE. V.et al. (2022). Morphological characteristics of spoken language in schizophrenia patients - an exploratory study. Scand. J. Psychol.63, 91–99. doi: 10.1111/sjop.12790

  • Pubmed Abstract
  • CrossRef
  • Google Scholar