Being able to locate salient stimuli in the visual field and paying attention to them is crucial to act in the world. Thus, the way visual search is organized is a key factor both in simple and complex daily life activities, such as arranging objects on a table or driving. Patients with left visual neglect, after right hemisphere lesions, show asymmetrical and inefficient visual search (Bartolomeo, 2014). For example, when asked to cancel multiple targets arranged in space, neglected patients typically omit to cancel left-sided targets. Their path of spatial exploration can be considered a broken line that links each marked target to the other. Whereas neurotypical subjects follow a more ordered and regular cancelation path, patients tend to adopt irregular exploratory patterns (Weintraub and Mesulam, 1988; Gauthier et al., 1989; Donnelly et al., 1999; Woods and Mark, 2007; Warren et al., 2008).

Based on this evidence, Mark et al. (2004) developed three indexes to quantify the spatial organization in a star cancelation task on patients with stroke:

To test the validity of these indexes, Woods and Mark (2007) administered to 50 adult observers 50 plots that reconstructed just as many pathways in a cancelation task. Participants assessed on a scale from 0 (= most organized) to 10 (= least organized) the overall organization of each path. All three Mark’s indexes (marking distance, Best r, and number of intersections) were found to be associated with observers’ scores or subjective organization.

Other authors tried to quantify the visual organization in cancelation tasks using these indexes (Ten Brink et al., 2016a,b, 2018; Benjamins et al., 2019) or tried to develop other measurements (Dalmaijer et al., 2015). Among them, it is worth mentioning

These and other indexes were applied in a recent study (Argiuolo et al., 2022), whose aim was to generalize these organization indexes to a new tool, the Enhanced Baking Tray Task (E-BTT). The E-BTT (Cerrato et al., 2019a,b, 2020; Gentile et al., 2019; Palumbo et al., 2019; Argiuolo and Ponticorvo, 2020; Somma et al., 2020; Argiuolo et al., 2021a) exploits tangible interfaces and has been developed based on the baking tray task, a widely used neglect assessment tool (Tham and Tegnér, 1996). The E-BTT is made up of 16 disks and a modular wooden frame (60 cm × 45 cm) equipped with tags (ArUco Markers, Garrido-Jurado et al., 2014), which are detected by a camera placed above the table, attached to a metallic arm (refer to Figure 1). Participants should place the disks, one by one, inside the frame as evenly as possible “as they were cookies to be baked in the oven.” The order is not predefined, but once placed, the disk can no longer be moved. A software platform, called E-TAN, processes the positions of the disks and returns a set of 16 ordered coordinates as output (Cerrato et al., 2020).

The fundamental difference with the cancelation task is the absence of predefined and therefore correct answers: the disks can be placed potentially anywhere and in any order, not decided a priori. This difference may trigger different spatial processes: if in the cancelation, it is essential to pay attention to the targets (and discriminate them from the distractors); in the E-BTT, each participant has an empty space in front of them in which they have to organize and fill in with the disks, knowing that they cannot be moved once they are placed. Moreover, the E-BTT can be considered ecologically valid, inasmuch as it refers to an everyday activity such as disposing of objects, and has a somatic component since it requires physically grabbing each disk and placing it inside the frame. It is more and more clear the importance of preserving the interaction with tangible objects in neuropsychological assessment. More awareness of the resulting bodily and situated aspects of spatial behaviors is needed (Pfeifer et al., 2007; Clark, 2008).

However, despite the differences, the organization of visual exploration should be crucial to process the space in the E-BTT, too. Indeed, the E-BTT needs the participant to organize the framed space, using it as much as possible and using all the available disks. Also, the sequence that goes from the first to the last-placed disk can be considered, which will result in a broken line just like in cancelation tasks pathways.

The abovementioned indexes (inter-disks distance, intersection rate, etc.) were proven to be usefully applied to the E-BTT (Argiuolo et al., 2022). Moreover, their structure was studied through a principal component analysis, and three out of the five latent components that emerged could be linked to the quality of the sequence itself. These latter indexes were then used in this article, starting from the hypothesis that they measure the organization/disorganization of the spatial configuration of the disks in the E-BTT. In other words, we expect that these indexes are significantly associated with the subjective judgment of external observers in terms of organization/disorganization, as Woods and Mark (2007) have already noted for the cancelation test.

Due to non-normal distributions, non-parametric analyses were preferred. The subjective organization measure’s median for each item (plot) was used in Spearman’s ρ to evaluate convergence. Table 1 summarizes the most important descriptive statistics for each index and the subjective organization measure.

The Spearman’s ρ correlations (Table 2) showed a significant association between subjective organization ratings and Intersection rate (r = –0.77, P < 0.05), correlations of each coordinates series and their order (respectively R_X: r = 0.26, P < 0.05 and R_Y: r = –0.36, P < 0.05), total time of performance (r = –0.3, P < 0.05), and distance to both optimal sequences (r = –0.36, P < 0.05 and r = –0.33, P < 0.05).

In order to analyze the eventual collinearity between these indexes, we also performed a correlation analysis (Table 3). All significant correlations were below 0.4, except the correlation between R_X and OS2_BD and the correlation between R_Y and OS1_BD.

We then inserted these indexes in a multiple regression model in order to obtain the relation between each of them and the subjective organization. Results showed that intersection rate, total time of performance, and distance to both optimal sequences are significantly related to the subjective evaluation of the organization, while R_X and R_Y’s estimates were not statistically significant (Table 4). The model explained about 48% of the variance (adjusted R² = 0.48) and the root mean square error was equal to 17.09. The collinearity statistics (Table 5) showed that there was no concern about collinearity because all variance inflation factor coefficients are below 5.

In Figure 4, two example sequences are shown. These two examples were chosen based on their subjective overall ratings. The first, in Figure 4A, is identified as well organized with high subjective ratings, no intersection, and a low distance from one of the two optimal sequences. The second, instead, has a high intersection rate, low subjective rating, and high distance from both optimal sequences. These can be considered two extremes in the subjective organization–disorganization continuum.

In this article, we explored how some indexes, recently applied to the E-BTT, could measure the organization of the spatiotemporal sequence output of the task. To do so, a subjective evaluation of each pattern’s overall organization was assessed. No information on the indexes was given to participants, but there was, however, a detailed description of the E-BTT and how it worked in order to let the participants familiarize themselves with the task.

Our initial hypotheses were only partially confirmed by results because not all the indexes included in this study were significantly associated with the subjective evaluation of spatial organization. From these results, it can be concluded that a plot was considered organized if

The intersection rate correlated negatively with the subjective evaluation scores of the overall organization because a higher number of intersections in the sequence were associated with lower ratings of the organization. The intersection rate has been widely used to measure search organization (Mark et al., 2004; Ten Brink et al., 2016a,b, 2018; Benjamins et al., 2019) and was associated with the activation of posterior regions of the right hemisphere in patients with stroke (Ten Brink et al., 2016a). Thus, disorganized search patterns seem to be associated with brain regions dedicated to conjunctive search and spatial working memory rather than with prefrontal regions (responsible for executive functions).

The amount of time a participant employs to finish a task can be considered a measure of executive functioning (Dalmaijer et al., 2015) because it reflects the ability to use one’s attentive resources throughout the whole task. The fast performance of the E-BTT may depend on a proper plan of action, resulting in an organized sequence of placement. Consistently, performance time was negatively associated with the subjectively judged organization. In other words, the plots judged as more organized were those completed faster.

Interestingly, the correlation between the Y-coordinates and their order is negatively associated with the overall subjective organization, whereas it is positively correlated to R_X. These two results were not confirmed when these two indexes were inserted in a multiple regression model, though. This could be explained by the high correlation between R_X and OS2_BD and between R_Y and OS1_BD which probably made spurious the relationship between R_X, R_Y, and the subjective evaluation.

Optimal sequences between distance is the Euclidean distance from each of the two sequences considered optimal. Our data suggest that the more a sequence is distant from these two sequences, the more disorganized they are evaluated. This could mean that if the sequence is very different from one of the two optima, it can be considered disorganized, thus implying that the two optimal sequences are organized.

In our study, WD (similar to inter-cancelation distance) was not associated with subjective evaluation, unlike in previous studies (Woods and Mark, 2007). This could mean that in E-BTT the intra-disks distance is not an organization index, but rather an index of how much space is used in the task, like the total area (Argiuolo et al., 2022). In our previous study, however, WD could actually discriminate between organized and disorganized plots (Argiuolo et al., 2021b), so further studies are needed to solve this issue.

Speed, standardized angle, and total area were not associated with the evaluation of the plot organization. Thus, perhaps the organization’s judgments did not take into substantial account the amount of space used by the sequence. When asked about their concept of “organization,” many refer to “order of the sequence,” “symmetry,” “sequentiality of numbers,” etc. Indeed, the observers reported having relied on “order,” “symmetry,” and “linearity” to judge each sequence as they report in the last question.

This study is not without limitations. First, we used a subjective measure of the overall organization, with all the limitations that a subjective measure entails (social desirability, response set, fatigue, etc.). In addition, participants were asked to judge the plots for their “overall organization” without a clear operationalization of the construct “organization” and thus relying on everyone’s subjective concept. It is also true that we followed Woods and Mark’s (2007) example, and they did not give participants a clear definition of “organization” either.

Secondly, the measure of subjective organization could be considered a measure of convergent validity, part of the construct validity (Rosenthal and Rosnow, 1991). Discriminant validity was not considered in this paper, so future studies should address discriminant validity in organization indexes in E-BTT.

Moreover, the plots we used referred all to healthy participants (Somma et al., 2020), while Woods and Mark (2007) also used reconstructed paths from patients with stroke. It would be interesting to test whether there were any differences in terms of organizational measures in the E-BTT between patients and healthy participants.

In conclusion, total time, distance from optimal sequences, and intersection rate can be considered indexes of spatial organization in the E-BTT task. Future studies should deepen the issue, distinguishing different neuropsychological syndromes resulting from different brain lesions. In parallel, the use of these indexes for research on healthy participants could help clear the relationship between spatial abilities and other important psychological variables, such as stress (Somma et al., 2021).

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Ethical Committee of Psychological Research, University of Naples Federico II, Department of Humanistic Studies. The patients/participants provided their written informed consent to participate in this study.

OG, MP, and PB contributed to the conception and design of the study. AA and FS collected the data. AA wrote the first draft and performed the statistical analysis. All authors contributed to the manuscript revision, read, and approved the submitted version.