Edited by: Sharlene D. Newman, Indiana University Bloomington, United States
Reviewed by: Donatella Rita Petretto, University of Cagliari, Italy; Ana Pinheiro, University of Lisbon, Portugal
This article was submitted to Cognition, a section of the journal Frontiers in Psychology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Semantic processing underpins the organization of verbal information for both storage and retrieval. Deficits in semantic processing are associated with both the risk for and symptoms presented in schizophrenia. However, studies are mixed and could reflect the confounding effects of medication and symptom heterogeneity. Therefore, we considered whether two risk phenotypes, positive schizotypy and hallucinatory predisposition, present in the general population were associated with differential responding profiles for a semantic processing task. One hundred and eighty-three participants completed the Schizotypal Personality Questionnaire, Launay-Slade Hallucination Scale, National Adult Reading Test, a handedness measure, and a computerized semantic relatedness judgment task. Pairs of words were related through their dominant or subordinate meanings, or unrelated. Participants were divided into four groups using a mean split on cognitive-perceptual (positive) schizotypy and hallucination proneness. Significant differences between groups were found for reaction time on the semantic relatedness task, with the high cognitive-perceptual schizotypy groups responding significantly slower to all word pairs compared to their low scoring counterparts. There was some evidence that high hallucination proneness was associated with significantly faster reaction times which may reflect disinhibitive processes, however additional support is required. The results suggest that these two components of psychosis risk are associated with different patterns of responding to semantic processing. More diffuse activation of semantic information appeared to be associated with positive schizotypy, while those predisposed to hallucinations appeared to respond quicker. These results have significant implications in the re-conceptualization of hallucination proneness as distinct from positive schizotypy. Additional research is required to investigate the association between psychotic-like experiences separate from personality variables such as positive schizotypy and semantic processing.
Semantic processing is the cognitive consideration of word meanings, where review of a particular word automatically stimulates activation for other words with similar and related meanings. Semantic processing abnormalities are a central feature of schizophrenia (
With schizophrenia at the extreme end, schizotypy exists along a continuum of psychosis (
Semantic priming is one of the main tasks used to investigate semantic networks. A priming word is presented (e.g., ball), followed by a target word which is semantically related (e.g., soccer) or unrelated (e.g., coffee). Semantic priming occurs when the participant responds to the related word significantly faster/more accurately than the unrelated word. The facilitation for the related word occurs because activation spreads across links in the network and semantically related nodes are located closer together, whilst unrelated words are further apart. Indirect priming occurs when the prime and target are not directly related, but mediated by another concept (e.g., prime CAT and target CHEESE are mediated by MOUSE). Indirect priming reflects a less constrained, or more diffuse, spread of activation (
Positive schizotypy is a complex trait comprising unusual beliefs, thoughts and perceptual experiences. Schizotypy is a trait that is relatively stable over time and reflects a complex component of personality, underpinned by an interaction between environment and genetic predisposition (
One mechanism proposed for the differential processing of ambiguous relations in high schizotypes is reduced cognitive inhibition; a process regarded as central to high schizotypy and directly related to language processing ability (
It is hypothesized that our groups will not differ for reaction time responses to dominant meaning related word pairs, and will be characterized by significantly faster reaction times to targets related by the dominant meaning of the prime compared to unrelated targets. For subordinate meaning word pairs, it is expected that the low positive schizotypy and AVH prone groups would exhibit significantly slower reaction times compared to dominant pairs, due to the inhibition of subordinate meanings (the expected semantic function in the general population;
One hundred and eighty-three undergraduate students from the University of Wollongong, NSW, Australia, took part in the study [mean age 22 years (SD 7.16), age range 17–60 years, 75.4% female] and were recruited on an opportunity sample basis.
Each participant completed an initial demographic questionnaire, requiring details such as age and sex. Handedness was determined, then participants completed the Schizotypal Personality Questionnaire (SPQ;
The SPQ consists of 74 items requiring either a yes or no response summed to produce a total score and three dimensions: Cognitive Perceptual (CP), Interpersonal, and, Disorganized. The overall mean score for participants on the SPQ was 27.28 (S.E. 1.18). The focus of the current study is positive schizotypy so participants were divided into groups based on their CP score. The mean score for participants on CP was 10.57 (S.E. 0.52). Those with a score above the mean were the high CP schizotypy group, while those scoring at or below the mean were the low CP schizotypy control group.
The LSHS is a 12-item questionnaire captures hallucinatory predisposition with a total score being calculated with a possible range of 0–12. The overall mean score for participants was 3.43 (S.E. 0.17). Those with a score above the mean were grouped as the high AVH prone group, whilst those scoring at or below the mean were the low AVH prone group.
Handedness was determined by the hand participants preferred to use when performing nine various tasks (i.e., writing, sweeping with a broom, unscrewing the lid of a jar). If Right/Left hand was used for 7+ activities, this determined handedness. However, if the Right/Left hand was used for less than seven activities, the individual was classified as “Mixed” handedness.
The NART was used to estimate verbal intellectual ability. Participants read aloud 50 atypically spelled words increasing in difficulty. The number of pronunciation errors was recorded.
In the present study, semantic processing was evaluated via the use of homographs (words with two different meanings). Disambiguating meaning in the English language requires the activation of the appropriate semantic pathway, and deactivation (inhibition) of the incongruous alternate meaning(s). In the current relatedness judgment task [from
This task was conducted on a laptop in a quiet room at the University of Wollongong. Participants were told they would see one word flash up on the screen (prime), followed immediately by another word (target). Once the target disappeared they were required to indicate whether both words were related (pressing key 1) or unrelated (pressing key 2). The task comprised 144 trials, participants were asked to respond as accurately and quickly as possible.
Forty different prime words were used, along with a target word related to the prime by the dominant meaning and a target related by the subordinate meaning the prime. For example, for the prime “ball,” a dominant related target would be “round,” whereas a subordinate related target would be “dancing.” Seventy-two related word pairs were used in the task (36 dominant and 36 subordinate), along with 72 completely unrelated word pairs (where a prime was pseudo-randomly paired with one dominant and one subordinate word of a different unassociated prime). Each participant saw each prime twice: paired with either a related target (half of which were dominant, half subordinate), or an unrelated target (half of which were dominant, half subordinate). Participants also saw each target on two occasions. Counterbalancing occurred, such that if a participant viewed a prime paired with a dominant and related target, they would also see the same prime paired with a subordinate and unrelated target, and vice versa. Counterbalancing also occurred across pairings with the use of two word lists. These word lists comprised the same words, however paired differently. For example, if in the first word list the related word pair was subordinate and the unrelated word pair was dominant, this would be reversed in the second word list (so the related word pair would be dominant and the unrelated word pair would be subordinate). Participants completed the task with word list 1
Each trial was preceded by a fixation mark in the center of the screen (1000 ms), followed by a centrally presented prime word (50 ms). A stimulus onset asynchrony (SOA) of 750 ms then followed, after which the target was presented for 180 ms. A SOA of 750 ms was used because this SOA is when inhibitory processes are most likely occurring (
Ethical approval to commence the study was obtained by the University of Wollongong Human Research Ethics Committee. Written informed consent was obtained from participants before testing commenced. Participants were reimbursed with course credit for their time.
Data analysis was conducted using SPSS Version 21 (
Performance accuracy was divided into two components: sensitivity (
The Signal Detection variables were calculated using the
To allow comparison of
In cases where the number of hits or false alarms was 0 or 1, an adjustment was applied to avoid infinite values. Proportions of 0 and 1 were converted using the formula 1/(2N) and 1-1/(2N), respectively, where N symbolizes the number of trials that proportion is based upon (
Main effects between the CP schizotypy groups revealed significant differences between high and low CP schizotypy groups for the SPQ Cognitive-Perceptual dimension [
Significant main effects were also found between high and low AVH proneness groups for SPQ total score [
Demographic variables for the interaction between CP schizotypy and AVH proneness are in
Demographic variables for interaction between Cognitive-Perceptual (CP) schizotypy groups and Auditory-Verbal Hallucination (AVH) proneness groups.
Variable | High CP schizotypy, High AVH prone ( |
High CP schizotypy, Low AVH prone ( |
Low CP schizotypy, High AVH prone ( |
Low CP schizotypy, Low AVH prone ( |
Test statistic and |
Sex (Male: Female) | 16:43 | 9:19 | 6:21 | 13:55 | χ2 = 2.254, |
Age | 21.98 (7.85) | 20.07 (3.54) | 21.26 (4.76) | 23.01 (8.33) | |
Handedness (Right:Left: Mixed) | 56:3:0 | 25:3:0 | 21:5:1 | 58:10:0 | χ2 = 10.322, |
SPQ total score | 42.75 (11.55) | 32.5 (11.63) | 21.48 (7.75) | 14.04 (9.42) | |
Interpersonal | 17.61 (6.16) | 14.21 (8.13) | 9.26 (4.94) | 6.88 (6.25) | |
Disorganized | 9.27 (3.57) | 6.25 (3.52) | 6.26 (3.72) | 3.46 (3.45) | |
NART total score | 28.59 (5.38) | 26.54 (5.06) | 29.15 (5.16) | 27.75 (5.28) |
Concordant response times were analyzed in a 2 (meaning) × 2 (relatedness) × 2 (CP schizotypy group) × 2 (AVH prone group) Repeated Measures ANOVA. In this design meaning and relatedness were the within subject variables, and CP schizotypy and AVH proneness were the between subject variables. All variables met the ±2 requirements for skewness and kurtosis except for the unrelated subordinate reaction time variable, which had a kurtosis value of 3.01 (SE 0.36). As a result box plot diagrams were used to identify possible outliers. One outlier was identified and removed, with the renewed kurtosis value subsequently meeting acceptable limits. Sphericity was not violated for this data therefore no corrections were required. When
Mean of the median reaction times to concordant responses in milliseconds.
Group | Meaning | Related | Unrelated | Unrelated – Related |
High CP Schizotypy, | Dominant | 826 (22) | 1077 (26) | 251 (25) |
High AVH prone | Subordinate | 987 (19) | 1062 (28) | 75 (23) |
High CP Schizotypy, | Dominant | 779 (18) | 1089 (32) | 309 (23) |
Low AVH prone | Subordinate | 998 (22) | 1116 (35) | 118 (29) |
Low CP Schizotypy, | Dominant | 697 (15) | 990 (24) | 294 (27) |
High AVH prone | Subordinate | 874 (19) | 958 (28) | 85 (29) |
Low CP Schizotypy, | Dominant | 770 (19) | 966 (21) | 196 (20) |
Low AVH prone | Subordinate | 936 (18) | 933 (20) | −3 (20) |
Main effects of both meaning [
A significant main effect was found for CP schizotypy group [
No main effect was found for AVH proneness (
No interaction was documented between CP schizotypy and AVH proneness (
Mean of the median reaction times (milliseconds) to related and unrelated word pairs.
High AVH prone |
Low AVH prone |
|||
Related | Unrelated | Related | Unrelated | |
High CP schizotypy | 906 (24) | 1070 (33) | 889 (35) | 1103 (48) |
Low CP schizotypy | 785 (35) | 974 (48) | 853 (22) | 950 (31) |
A measure was calculated by subtracting the “related response time” from the “unrelated response time” for each participant. Using correlations we found response times between dominant and subordinate word pairs were significantly positively correlated (High CP schizotypy, High AVH prone [
An analysis with items as cases was used to confirm the results obtained by the previous by-subject (
There was a significant effect of relatedness [
A significant main effect was found for CP schizotypy group [
The main effect of AVH proneness was significant [
No interaction was found between CP schizotypy and AVH proneness (
Median reaction time (RT), response (in milliseconds) to relate word pairs. Line indicates Cognitive-Perceptual (CP) schizotypy group with responses broken down according to auditory verbal Hallucination (AVH) proneness.
Congruity across
Task effects of meaning, relatedness, and their interaction;
Group effects of CP schizotypy, indicating slower response times of those in the high CP schizotypy group;
Significant interaction between CP schizotypy, AVH proneness and relatedness, indicating faster responses of the low CP Schizotypy/High AVH prone group to related words.
However, the main effect of the high AVH proneness group for response times, and the interaction of CP schizotypy with meaning were only significant in the
Next the data were considered using signal detection analyses (see
Mean values for sensitivity and relatedness judgments in Cognitive-Perceptual (CP) schizotypy and Auditory Verbal Hallucination (AVH) prone groups.
Group | Meaning | % Hits | % False alarms | ||
High CP Schiz, | Dominant | 82.7 (14) | 13 (10) | 2.25 (0.71) | 0.03 (0.42) |
High AVH prone | Subordinate | 62.2 (14) | 1.52 (0.54) | ||
High CP Schiz, | Dominant | 84.1 (14) | 16.8 (14) | 2.13 (0.94) | −0.17 (1.04) |
Low AVH prone | Subordinate | 62.6 (14) | 1.39 (0.67) | ||
Low CP Schiz, | Dominant | 82.9 (16) | 12.3 (8) | 2.35 (0.78) | −0.22 (1.4) |
High AVH prone | Subordinate | 63.9 (16) | 1.59 (0.59) | ||
Low CP Schiz, | Dominant | 78.2 (19) | 14.8 (12) | 2.27 (0.96) | 0.23 (1.57) |
Low AVH prone | Subordinate | 57.6 (15) | 1.48 (0.67) |
The sensitivity (
Results indicated that all groups use the same criterion regardless of stimuli type. Mean values were: High CP schizotypy = 1.2 (SD 0.52), Low CP schizotypy = 1.19 (SD 0.47), High AVH prone = 1.25 (SD 0.46), Low AVH prone = 1.16 (SD 0.43).
To compare
The current study considered whether people differed in their performance on a semantic processing task according to their scores on hallucinatory proneness and positive schizotypy. The semantic task primed words according to their dominant or subordinate meaning in an attempt to see whether participants differed in their speed to construe meaning or relations between the words presented. Therefore, participants were divided into groups according to their state hallucinatory predisposition or trait positive schizotypy. This state-trait distinction may be considered debatable by some readers. The LSHS asks people to indicate whether they have had different types of perceptual experiences, moving from vivid imagery through to more fully formed hallucinations (
Results indicated those who were high on CP schizotypy responded significantly slower than those low on CP schizotypy. Contrastingly, some evidence was found for those predisposed to hallucinations to respond to word pairs faster than their respective low scoring counterparts. These findings are indicative of disparities in how state (AVH) and trait (schizotypy) psychosis risk factors influence processing of semantic relations. The slower overall response speed associated with CP schizotypy suggests increased difficulty in the processing of semantic information. It may be that in trait schizotypy, a diffuse spread of semantic activation results in more semantic nodes being activated (
In contrast, the relatedness effects demonstrated by the high AVH prone group in one (but not both) reaction time analyses suggests disinhibited processes may be contributing to significantly faster task completion. In non-clinical AVH samples, the tendency to jump to conclusions and interpret an internally generated experience as a true sensory experience has been suggested as a central mechanism in the generation and maintenance of hallucinations (
Although not predicted, compared to those with low CP schizotypy and low AVH proneness, those with high CP schizotypy and high AVH proneness responded to related word pairs significantly slower, whilst those with low CP schizotypy and high AVH proneness responded to related word pairs significantly faster. This interaction suggests that there may be two mechanisms at work. CP schizotypy appears to result in a more diffuse spread of semantic activation, which slows response times to related word pairs. Contrastingly, AVH proneness seems to reflect disinhibitive processes, such that relationships between semantic associates are responded to significantly faster as long as schizotypy is low/normal. These findings indicate that high CP schizotypy has a far more influential effect on the atypical processing of semantic relations than AVH proneness. These findings suggest that hallucination proneness is separable from positive trait schizotypy. Such a finding is in line with previous research (
A priming measure was also calculated for each participant for both dominant and subordinate words and, across all groups, correlations were similar in magnitude. Although the speed of processing differs between groups, the current study suggests that the organization of the semantic system is similar, at least for normatively associated words. These finding suggest that scoring highly on CP schizotypy or AVH proneness has no effect on the ability to detect relationships between words. In addition, no significant differences were found between the CP schizotypy and AVH prone groups for signal detection outcomes. Research suggests high schizotypes require additional task demands before the breakdown in control processes (such as inhibition) that organize semantic processing occurs (
A number of limitations need to be taken into account. The sample consisted of reasonably high functioning university students who generally have higher cognitive, social, and often financial resources compared to community samples. Consequently the failure to find significant differences in signal detection criteria may be the result of the current sample not being representative of the spread of ability in the general population. However, high error rates on the NART suggest we did have a wide spread of verbal ability. Furthermore, the current study used the CP schizotypy factor to split high and low schizotypal groups. Although this has been used in previous studies testing for semantic processing abnormalities (
In conclusion, this study considered the nature of semantic processing disturbances in both high trait CP schizotypy and high state AVH prone groups. Our findings indicate that the speed of processing ambiguous semantic relations varies according to level of trait and state psychosis risk. From these initial comparisons, it appears that the slower speed of semantic processing found in high CP schizotypy may be related to a more diffuse spread of semantic activation. Contrastingly the semantic processing capabilities associated with AVH proneness may be related to disinhibitive processes, resulting in an accurate and efficient speed of decision making for semantic information, but only in the context of low CP schizotypy. Previously, positive schizotypy and AVH proneness were believed to be synonymous indications but our study suggests further investigation is required to determine the separation between these two phenotypes.
The datasets generated for this study are available on request to the corresponding author.
The study was reviewed and approved by the Social Sciences and Humanities Human Research Ethics Committee at the University of Wollongong. The participants provided written informed consent to participate in this study.
SL-S was involved in the study design, led the data collection, wrote the first draft of the manuscript, and performed the analysis. SR and EB contributed to the study design, analysis, and were involved in writing the manuscript. LH, SM, and EM were involved in the data collection and proofread and contributed to the manuscript. All authors contributed to the article and approved the submitted version.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.