Background: Impairments in theory of mind (ToM) are highly prevalent among individuals with schizophrenia, resulting in substantial functional deficits. However, research on impairments in individuals with schizotypy has yielded inconsistent findings, with some studies finding ToM deficits in overall schizotypy, other studies finding ToM deficits in only specific schizotypy dimensions, and yet other studies finding no ToM deficits at all. One potential key factor that may account for this discrepancy is the use of schizotypy measures that do not adequately measure specific schizotypy dimensions. Additional limitations are employment of ToM measures that rely heavily on explicit cultural knowledge, verbal/reading comprehension, and/or other cognitive abilities. Method: To address these discrepant findings, we used the Schizotypal Personality Questionnaire-Brief Revised (Updated; SPQ-BRU) and the Multidimensional Schizotypy Scale (MSS) to tap overall schizotypy and specific schizotypy dimensions. To measure ToM, we used the Frith-Happé animations (FHA) and Strange Stories Film Task (SSFT). We examined the hypothesized negative relationship between schizotypy and ToM in a sample of 233 nonclinical individuals. Results: Regression analysis indicated no significant relationship between overall schizotypy and ToM on both the FHA (b = 0.01, t(196) = −0.75, p = 0.46) and SSFT (b = −0.20, t(195) = −1.69, p = 0.09). However, it did find that the negative schizotypy dimension was associated with poorer ToM performance on both the FHA (b = −0.11, t(194) = −2.7, p = 0.008) and SSFT (b = −0.12, t(193) = −3.22, p = 0.001). Also, exploratory analyses employing an extreme-group design approach indicated high schizotypy and high negative schizotypy groups displayed weaker ToM performance within all specific schizotypy dimensions. Conclusion: These results indicate that ToM impairments are present in schizotypy, especially within the negative schizotypy dimension. The results suggest important methodological implication for studying ToM in schizotypy and conceptualizing the latent structure of schizotypy.

Schizotypy has been defined as a spectrum of experiences and behaviors ranging from subclinical abnormalities to severe mental illness that encompasses subclinical expressions, prodromal manifestations, schizophrenia-related personality disorders, and schizophrenia spectrum disorders such as delusional disorder, schizoaffective disorder, and schizophrenia itself [1], with schizophrenia considered the most extreme expression of a spectrum of impaired functioning [2]. While not considered a diagnostic category, schizotypy is viewed as a latent personality organization that harbors the vulnerability for schizophrenia and other related disorders [3].

The work of Meehl [4] and Lenzenweger [2] developed schizotypy construct to reflect the diathesis stress model of schizophrenia such that schizotypy refers to the genetic vulnerability to develop schizophrenia and related disorders, depending on interaction with the environment. According to the schizotypy construct, schizotypic individuals may not display any observable behavioral abnormalities and instead must be detected using psychometric methods [2]. It is important to distinguish the schizotypy construct from several other related but distinct constructs, such as schizotypal personality disorder, psychosis-proneness [5], and the clinical high-risk approach [6]. For example, despite potential overlap in phenomenology across these constructs and schizotypy, a key distinguishing factor is that the schizotypy model does not identify schizotypic individuals based on whether they meet DSM-specific diagnostic criteria based on clinical observation (e.g., schizotypal personality disorder) or display significant behavioral or cognitive impairments (clinical high risk). In addition, while schizotypy allows researchers to identify schizotypic individuals to explore potential impairments and treatment targets, it is less concerned with predicting conversion to psychosis as is the case with the psychosis-proneness model.

Like schizophrenia, schizotypy is a multidimensional phenomenon that is comprised of positive, negative, and disorganized dimensions [2, 3]. The positive schizotypy dimension includes disruptions in thought content, ranging from magical ideation to delusional beliefs; perceptual abnormalities, e.g., hallucination; and suspiciousness/paranoia. The negative schizotypy dimension includes deficits in functioning and experience, e.g., alogia, anhedonia, avolition, flattened affect, diminished emotional processing, asociality. The disorganized schizotypy dimension includes disturbances in the organization of thought and behavior, e.g., ranging from slight abnormalities to formal thought disorder and severely incoherent speech and actions. The schizotypy construct provides researchers with the ability to examine the etiology and expression of schizophrenia spectrum psychopathology without confounds like medication effects and hospitalization [7]. Numerous studies have shown that evaluating impairments using psychometrically defined schizotypy in nonclinical samples has tapped into similar deficits in disorders along the clinical diagnostic range of schizotypy, including schizophrenia (for a review, see [3]). Investigating such deficits in nonclinical samples can inform treatment for psychotic disorders and guide early intervention.

Among the factors contributing to functional impairment in psychotic disorders, impaired social cognition, and specifically theory of mind (ToM), contributes significantly to the interpersonal deficit in psychotic disorders and is a potential target for intervention [8, 9]. ToM refers to the ability to correctly infer mental states to oneself and others [10]. ToM is often used synonymously with mentalization, which is the broader capacity to understand self and other behavior in terms of mental states such as beliefs, desires, intentions, feelings, and emotions [8]. Investigating potential ToM impairments in nonclinical schizotypy samples can inform treatment for psychotic disorders and guide early intervention. Further, studying schizotypy in a college-age population is important because early adults are at increased risk for developing psychotic disorders and is thus a crucial intervention period [9].

While numerous studies of ToM in schizotypy have found significant associations between general ToM impairment and schizotypy (e.g., [11, 12]), others have found ToM deficits in only specific dimensions [13, 14]. Limitations of these studies include small sample size (e.g. [12, 13]), the theoretically unsupported use of median-split or cutoff scores to divide samples [12, 14], the use of outdated schizotypy measures [13], and the use of limited ToM measures that do not tap the full range of ToM and that might be confounded by verbal comprehension or cultural knowledge [11, 12].

One key factor that may account for the discrepancy in findings is the way in which schizotypy is defined. While there are several conceptualizations of schizotypy, a core commonality to most views is that schizotypy represents a neurodevelopmental vulnerability for schizophrenia across a continuum ranging from minimal impairment to schizophrenia [1]. A “fully dimensional” model of schizotypy views schizotypy as a continuous spectrum from normal to pathological that is normally distributed in the population [15]. In contrast, the taxonic model of schizotypy views schizotypy as a spectrum of abnormal traits ranging from subclinical manifestations to schizophrenia exhibited by only a specific portion of the population, e.g., 10% of the population [2, 4]. Along with controversy regarding the latent structure of schizotypy, there is disagreement regarding whether scores on schizotypy measures should be studied continuously using regression-based approaches, or categorically using an extreme-group design (EGD) [16] paradigm using mean comparison methods that compare low and high schizotypy groups. The EGD approach in schizotypy research often stipulates some cutoff value on a schizotypy measure to sort study participants into low and high schizotypy groups, e.g., using a median-split method, percentile scores, or cutoff value based on one or two standard deviations (SDs) from the sample mean.

To address these limitations and gaps in the literature, this study had the following aims: (1) to examine the impact of schizotypy on ToM using the recently validated Multidimensional Schizotypy Scale (MSS [17]) to examine specific schizotypy dimensions, (2) to use two ToM measures that do not rely heavily on explicit cultural knowledge and verbal/reading comprehension or other cognitive abilities on ToM. To our knowledge, the present study is the first to use the MSS, the newly developed Strange Stories Film Task (SSFT [18]) and the Frith-Happé animations (FHA [18‒21]) together in a study of ToM in schizotypy. We hypothesized that (1) using regression analysis, higher schizotypy, both overall and within specific dimensions, would significantly predict poor ToM, and (2) using an EGD approach, high schizotypy groups would perform worse on ToM tasks compared to low schizotypy group.

Procedure

The current study was part of a larger study examining the impact of emotion regulation difficulties and negative effect on ToM in schizotypy. Participants were undergraduate students at a large, urban university in New York City recruited via a research pool comprised of Introductory Psychology students who were required to participate in research studies for course credit. The study was approved by the university’s Institutional Review Board (IRB), and all participants provided informed consent. Participants were included in the study if they were current undergraduates and were 18–30 years old. Participants completed the study procedures remotely online using Qualtrics. Upon completion, participants were presented with a debriefing form and awarded credit. All participants received course credit for completing the study.

Measures

Demographics Questionnaire

Participants provided demographic information including age, race, ethnicity, country of origin, gender, education, whether they had ever been diagnosed with a psychiatric disorder, and whether they had any family history of mental illness.

Schizotypy

  • 1.

    Overall schizotypy. The Schizotypal Personality Questionnaire-Brief Revised (Updated) (SPQ-BRU [22]) was used to measure overall schizotypy traits. The SPQ-BRU is a 32-item self-report measure of schizotypy that yields an overall score, which has shown sufficient reliability (α = 0.91), as well as three subscale scores, all of which have shown sufficient reliability: cognitive-perceptual (α = 0.85), interpersonal (α = 0.85), and disorganized (α = 0.86). SPQ-BRU asks participants to self-report on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree) to reflect agreement with statements related to schizotypy traits and experiences. The SPQ-BRU overall schizotypy scale showed good reliability in the present study (α = 0.90). Total SPQ-BRU scores were computed by summing participant responses, with higher sum scores representing higher levels of schizotypal personality traits.

  • 2.

    Specific schizotypy dimensions. The MSS [17] was used to measure specific positive, negative, and disorganized schizotypal traits. The MSS is a 77-item self-report measure of multidimensional schizotypy consisting of three subscales corresponding three dimensions of schizotypy, all of which showed good reliability: positive (α = 0.88), negative (α = 0.84), and schizotypy (α = 0.92). Participants answered true or false to reflect agreement with statements related to schizotypal traits/experiences (e.g., “I have sometimes felt that strangers were reading my mind.”). These variables were scored continuously from low to high with higher sum scores presenting higher respective levels of specific dimensional positive, negative, and disorganized schizotypy traits. The MSS subscales all demonstrated good reliability in the present study: positive (α = 0.88), negative (α = 0.84), and schizotypy (α = 0.92). Respective subscale scores for specific schizotypy dimensions were studied continuously and were used for exploratory analyses.

Theory of Mind

  • 1.

    The Frith-Happé animations (FHA [18‒20]). The FHA is a computer animation-based ToM measure. It has since been validated and used in studies of nonclinical schizotypy [13] and psychotic disorders [23]. The FHA is an ecologically valid assessment of ToM ability that measures “online” mentalizing without relying on other cognitive functions, verbal cues, or explicit cultural context. Participants viewed 12 short (∼45 s), computer-presented animations on a computer screen, which shows one large, red and one small, blue triangle moving around the screen. The FHA is comprised of two practice items and 10 test items. In some animations, the shapes move according to a scripted intentional mental interaction (e.g., persuading); in other animations, the shapes interact in goal-directed manner that does not require the manipulation of another’s mental state (e.g., dancing, fighting); finally, in other animations, the triangles move randomly without any scripted narrative. After viewing each item, participants were asked to categorize what they viewed as one of three script types, i.e., mental interaction, physical interaction, or no interaction, respectively. Correct answers on categorization of the mental interaction items received 1 point on the FHA-Categorization subscale (FHA-Cat). Correct answers yielded two follow-up questions about the feelings of the triangles according to the clips, which received one point on the FHA-Feelings subscale (FHA-Feel). The FHA-Cat (α = 0.74) and FHA-Feel (α = 0.77) subscales showed good reliability in the present study.

  • 2.

    The Strange Stories Film Task (SSFT [18]). The SSFT is an ecologically valid film-based assessment of ToM based on the original Strange Stories Task [21], which is a ToM measure that has been used in previous schizotypy studies (e.g., [13]). Participants were required to view 15 short computer-presented film clips on a computer screen involving different social scenarios. After each clip, participants were asked a question assessing intentionality (e.g., “Why did the character say that?”), interaction (e.g., “If you were the character, what would you say next?”), and a memory question (e.g., “Where was the character?”). For each question, participants were provided with four possible answer choices with one correct answer choice and three incorrect choices. Participants were asked to rate each answer choice on a scale from 0 (very unlikely) to 3 (very likely). Each SSFT item was scored according to a formula that weighed ratings of the correct answer choice against ratings of three remaining incorrect answer choices. Total sum scores for the SSFT were analyzed continuously. The SSFT showed good reliability in the validation study (α = 0.73) as well as in the present study (α = 0.78).

General Cognitive Functioning

General cognitive functioning (Cog. Fx) was measured using the International Cognitive Ability Resource (ICAR [24]), which is a brief 16-item measure that has been validated in a large online sample, showing good reliability (α = 0.81). The ICAR consists of four items of four question types: matrix reasoning, verbal reasoning, three-dimensional rotation, and letter-number sequences. Participants were asked to select correct answers from multiple-choice responses (e.g., the correct shape that completes a series or the answer to an arithmetic problem). The ICAR showed good reliability in the present study (α = 0.75).

Positive and Negative Affect

The Positive and Negative Affect Schedule (PANAS [25]) is a 20-item scale including 10 items related to positive affect and 10 items related to negative affect. Participants rated on a scale of 1 (very slightly or not at all) to 5 (extremely) the degree to which they are currently experiencing a series of affective experiences, such as “jittery,” “hostile,” or “enthusiastic.” This scale is widely used in undergraduate populations and has demonstrated sufficient reliability for the positive scale (α = 0.86–0.90) and the negative scale (α = 0.84–0.87). The PANAS positive affect (α = 0.91) and negative affect (α = 0.86) subscale scores demonstrated good reliability in the present study.

Valid Responding

The Jackson Infrequency Scale [26] is a commonly used 13-item scale that checks valid responding and attention while completing survey items by asking participants to answer true or false to reflect their agreement with statements about extremely strange or unusual beliefs and behaviors. Items from the scale were interspersed throughout the survey. Total scores were analyzed. Protocols containing 4 or more inappropriate answers were inspected for further indications of invalid responding.

Statistical Analyses

Data analyses were conducted using SPSS Version 28 [27]. Power analyses conducted using G*Power 3 [28] indicated that a sample size of 165 was needed to detect a medium effect size using multiple regression (f2 = 0.15), with an alpha of 0.05, a power of 0.8. Preliminary analyses were conducted to assess missing data, invalid protocols, internal consistency of study scales, normality of data. Correlation analyses were run to determine multicollinearity between variables. Potential covariates, such as age, years of education, explicit positive and negative affect, Cog. Fx, gender, ethnicity, and English as a first language, were analyzed and controlled for as needed in further analyses. For hypothesis testing, hierarchical multiple regression analyses and one-way ANOVAs were conducted for regression-based and EGD analyses, respectively.

Demographic Characteristics, Descriptive Statistics, and Covariates

A total of 233 individuals consented to participate in the study; 22 participants were excluded due to exceeding the maximum age (e.g., 30 years old) for study inclusion. In addition, 10 protocols were excluded due to having both high infrequency scores (>4) and at least one additional indication of invalid responding, such as poor effort (e.g., selecting the same Likert scale response for every question) and unexpected completion times, either excessively short (e.g., less than 20 min) or long (e.g., greater than 5 h, which suggests that a participant left and returned to the survey at a later time, despite study instructions to complete the study in one sitting). The average completion time of the study was approximately 3 h and 29 min.

The final sample size used for data analysis was 201 participants. Table 1 displays demographic data for the overall sample. Descriptive statistics for the primary study variables are presented in Table 2.

Table 1.

Sample characteristics

VariablenM (SD) or %
Age 201 19.55 (2.38) 
Sex 
 Female 145 72.1% 
 Male 54 26.9% 
 Nonbinary 0.5% 
 Prefer not to answer 0.5% 
Race 
 Western or Eastern European/White 49 24.4% 
 African American/Caribbean/Black 33 16.4% 
 South Asian 30 14.9% 
 Middle Eastern 26 12.9% 
 Hispanic/Latinx/Spanish Origin 20 10.0% 
 East Asian 15 7.5% 
 Southeast Asian 10 5.0% 
 Central Asian 2.5% 
 American Indian/Alaskan Native 1.5% 
 Native Hawaiian/other Pacific Islander 1.0% 
 Other 4.0% 
First language 
 English as a first language 138 68.7% 
 English as a second language 63 31.3% 
Lifetime psychiatric diagnosisa 21 10.4% 
VariablenM (SD) or %
Age 201 19.55 (2.38) 
Sex 
 Female 145 72.1% 
 Male 54 26.9% 
 Nonbinary 0.5% 
 Prefer not to answer 0.5% 
Race 
 Western or Eastern European/White 49 24.4% 
 African American/Caribbean/Black 33 16.4% 
 South Asian 30 14.9% 
 Middle Eastern 26 12.9% 
 Hispanic/Latinx/Spanish Origin 20 10.0% 
 East Asian 15 7.5% 
 Southeast Asian 10 5.0% 
 Central Asian 2.5% 
 American Indian/Alaskan Native 1.5% 
 Native Hawaiian/other Pacific Islander 1.0% 
 Other 4.0% 
First language 
 English as a first language 138 68.7% 
 English as a second language 63 31.3% 
Lifetime psychiatric diagnosisa 21 10.4% 

M, mean; SD, standard deviation.

aNo participants reported a diagnosis of a schizophrenia spectrum disorder or other psychotic disorder.

Table 2.

Descriptive statistics of study variables

nαM (SD)Skew (SE)Kurtosis (SE)
MSS-P 201 0.88 8.59 (6.00) 0.65 (0.17) −0.16 (0.34) 
MSS-N 201 0.84 5.71 (0.33) 0.62 (0.17) −0.77 (0.34) 
MSS-D 201 0.92 7.56 (6.50) 0.63 (0.17) −0.74 (0.34) 
SPQ-BRU 201 0.90 91.85 (18.59) −0.14 (0.17) 0.04 (0.34) 
DERS 201 0.94 90.61 (24.75) 0.21 (0.17) −0.26 (0.34) 
IPANAT-NA 201 0.84 1.90 (0.49) 0.38 (0.17) 0.00 (0.34) 
FHA-Cat 201 0.74 6.86 (2.65) −0.27 (0.17) −0.97 (0.34) 
FHA-Feel 201 0.77 2.63 (2.29) 0.53 (0.17) −0.75 (0.34) 
SSFT 201 0.78 5.19 (2.34) −0.50 (0.17) −0.69 (0.34) 
ICAR 201 0.75 6.93 (3.25) 0.21 (0.17) −0.49 (0.34) 
nαM (SD)Skew (SE)Kurtosis (SE)
MSS-P 201 0.88 8.59 (6.00) 0.65 (0.17) −0.16 (0.34) 
MSS-N 201 0.84 5.71 (0.33) 0.62 (0.17) −0.77 (0.34) 
MSS-D 201 0.92 7.56 (6.50) 0.63 (0.17) −0.74 (0.34) 
SPQ-BRU 201 0.90 91.85 (18.59) −0.14 (0.17) 0.04 (0.34) 
DERS 201 0.94 90.61 (24.75) 0.21 (0.17) −0.26 (0.34) 
IPANAT-NA 201 0.84 1.90 (0.49) 0.38 (0.17) 0.00 (0.34) 
FHA-Cat 201 0.74 6.86 (2.65) −0.27 (0.17) −0.97 (0.34) 
FHA-Feel 201 0.77 2.63 (2.29) 0.53 (0.17) −0.75 (0.34) 
SSFT 201 0.78 5.19 (2.34) −0.50 (0.17) −0.69 (0.34) 
ICAR 201 0.75 6.93 (3.25) 0.21 (0.17) −0.49 (0.34) 

n, number of participants; MSS-P, Multidimensional Schizotypy Scale, Positive Schizotypy subscale; MSS-N, Multidimensional Schizotypy Scale, Negative Schizotypy subscale; MSS-D, Multidimensional Schizotypy Scale, Disorganized Schizotypy subscale; SPQ-BRU, Schizotypal Personality Questionnaire-Brief Revised (Updated); DERS, Difficulties In Emotion Regulation Scale; IPANAT-NA, Implicit Positive and Negative Affect Test, Negative Affect subscale; FHA-Cat, Frith-Happé animations, Categorization subscale; FHA-Feel, Frith-Happé animations, Feelings subscale; SSFT, Strange Stories Film Task; M, mean; SD, standard deviation; SE, standard error.

Respective examination of potential covariates revealed determined age and Cog. Fx were included as covariates in analyses predicting FHA-Cat, FHA-Feel, and SSFT. In addition, explicit negative affect and ethnicity were included as covariates in models predicting SSFT and gender was included in models predicting FHA-Cat and FHA-Feel.

Hypothesis Testing Using Regression

Overall Schizotypy and ToM

It was hypothesized that there would be a significant, negative effect of overall schizotypy on ToM task performance, as measured by the FHA’s two subscales (FHA-Cat and FHA-Feel) and the SSFT.

Frith-Happé Animations (FHA). A hierarchical multiple regression analysis was conducted with age, Cog. Fx, and gender as covariates with overall schizotypy entered as the predictor variable and ToM, as measured by the FHA-Cat as the outcome variable (Table 3). The overall regression was significant, F(4,196) = 10.48, p < 0.001, R2 = 0.18. However, there was no significant relationship between overall schizotypy and ToM task performance (b = 0.01, t(196) = −0.75, p = 0.46). A similar null result was found to use the FHA-Feel as outcome variable (Table 4).

Table 3.

Hierarchical multiple regression results for FHA-Cat as dependent variable

BB (SE)βtp valueRR2R2
Step 1a      0.42 0.17 0.17 
 (Constant) 8.62 1.54  5.58 <0.001***    
 Age −0.11 0.07 −0.10 −1.55 0.12    
 Cog. Fx 0.28 0.05 0.36 5.51 <0.001***    
 Gender −0.63 0.28 −0.15 −2.23 0.03*    
Step 2b      0.42 0.18 0.00 
 (Constant) 7.89 1.82  4.33 <0.001***    
 Age −0.10 0.07 −0.10 −1.46 0.15    
 Cog. Fx 0.28 0.05 0.36 5.51 <0.001***    
 Gender −0.64 0.28 −0.15 −2.25 0.03*    
 SPQ-BRU 0.01 0.01 0.05 0.75 0.46    
BB (SE)βtp valueRR2R2
Step 1a      0.42 0.17 0.17 
 (Constant) 8.62 1.54  5.58 <0.001***    
 Age −0.11 0.07 −0.10 −1.55 0.12    
 Cog. Fx 0.28 0.05 0.36 5.51 <0.001***    
 Gender −0.63 0.28 −0.15 −2.23 0.03*    
Step 2b      0.42 0.18 0.00 
 (Constant) 7.89 1.82  4.33 <0.001***    
 Age −0.10 0.07 −0.10 −1.46 0.15    
 Cog. Fx 0.28 0.05 0.36 5.51 <0.001***    
 Gender −0.64 0.28 −0.15 −2.25 0.03*    
 SPQ-BRU 0.01 0.01 0.05 0.75 0.46    

Cog. Fx, cognitive functioning; SPQ-BRU, Schizotypal Personality Questionnaire-Brief Revised (Updated); FHA-Cat, Frith-Happé animations, Categorization subscale; B, unstandardized beta; B (SE), standard error of unstandardized beta; β, standardized beta; t, t test statistic; p, significance.

*p < 0.05, ***p < 0.001.

aF(3,197) = 13.82, p < 0.001, Adj. R2 = 0.16.

bF(4,196) = 10.48, p < 0.001, Adj. R2 = 0.16.

Table 4.

Hierarchical multiple regression results for FHA-Feel as dependent variable

BB (SE)βtp valueRR2R2
Step 1a      0.44 0.19 0.19 
 (Constant) 4.40 1.37  3.22 <0.01*    
 Age −0.13 0.06 −0.13 −2.07 0.04    
 Cog. Fx 0.26 0.05 0.37 5.71 <0.001***    
 Gender −0.60 0.25 −0.15 −2.40 0.02    
Step 2b      0.45 0.20 0.01 
 (Constant) 3.26 1.61  2.02 0.05    
 Age −0.12 0.06 −0.12 −1.91 0.06    
 Cog. Fx 0.26 0.05 0.37 5.73 <0.001***    
 Gender −0.61 0.25 −0.16 −2.44 0.02    
 SPQ-BRU 0.01 0.01 0.09 1.34 0.18    
BB (SE)βtp valueRR2R2
Step 1a      0.44 0.19 0.19 
 (Constant) 4.40 1.37  3.22 <0.01*    
 Age −0.13 0.06 −0.13 −2.07 0.04    
 Cog. Fx 0.26 0.05 0.37 5.71 <0.001***    
 Gender −0.60 0.25 −0.15 −2.40 0.02    
Step 2b      0.45 0.20 0.01 
 (Constant) 3.26 1.61  2.02 0.05    
 Age −0.12 0.06 −0.12 −1.91 0.06    
 Cog. Fx 0.26 0.05 0.37 5.73 <0.001***    
 Gender −0.61 0.25 −0.16 −2.44 0.02    
 SPQ-BRU 0.01 0.01 0.09 1.34 0.18    

Cog. Fx, cognitive functioning; SPQ, Schizotypal Personality Questionnaire-Brief Revised (Updated); FHA-Cat, Frith-Happé animations, Categorization subscale; B, unstandardized beta; B (SE), standard error of unstandardized beta; β, standardized beta; t, t test statistic; p, significance.

*p < 0.05, **p < 0.01, ***p < 0.001.

aF(3,197) = 15.70, p < 0.001, Adj. R2 = 0.18.

bF(4,196) = 12.27, p < 0.001, Adj. R2 = 0.18.

Strange Stories Film Task (SSFT). A hierarchical multiple regression analysis was conducted with age, explicit negative affect, and ethnicity as covariates with overall schizotypy entered as the predictor variable and ToM, as measured by the SSFT as the outcome variable (Table 5). The overall regression was significant, F(5,195) = 8.25, p < 0.001, R2 = 0.18. However, there was no significant relationship between overall schizotypy and ToM task performance after controlling for age, Cog. Fx, explicit negative affect, and ethnicity (b = −0.20, t(195) = −1.69, p = 0.09). The results indicate that overall schizotypy does not impact ToM, as measured by the SSFT.

Table 5.

Hierarchical multiple regression results for SSFT as dependent variable

BB (SE)βtp valueRR2R2
Step 1a      0.40 0.16 0.16 
 (Constant) 7.41 1.59  4.67 <0.001    
 Age −0.18 0.07 −0.17 −2.65 0.01    
 Cog. Fx 0.21 0.05 0.27 4.11 <0.001    
 PANAS– −0.04 0.02 −0.11 −1.63 0.11    
 Ethnicity 0.75 0.38 0.13 1.98 0.05    
Step 2b      0.42 0.18 0.01 
 (Constant) 8.94 1.82  4.92 <0.001    
 Age −0.19 0.07 −0.19 −2.85 0.01    
 Cog. Fx 0.21 0.05 0.28 4.21 <0.001    
 PANAS– −0.03 0.02 −0.07 −1.06 0.29    
 Ethnicity 0.69 0.38 0.12 1.82 0.07    
 SPQ-BRU −0.02 0.01 −0.12 −1.69 0.09    
BB (SE)βtp valueRR2R2
Step 1a      0.40 0.16 0.16 
 (Constant) 7.41 1.59  4.67 <0.001    
 Age −0.18 0.07 −0.17 −2.65 0.01    
 Cog. Fx 0.21 0.05 0.27 4.11 <0.001    
 PANAS– −0.04 0.02 −0.11 −1.63 0.11    
 Ethnicity 0.75 0.38 0.13 1.98 0.05    
Step 2b      0.42 0.18 0.01 
 (Constant) 8.94 1.82  4.92 <0.001    
 Age −0.19 0.07 −0.19 −2.85 0.01    
 Cog. Fx 0.21 0.05 0.28 4.21 <0.001    
 PANAS– −0.03 0.02 −0.07 −1.06 0.29    
 Ethnicity 0.69 0.38 0.12 1.82 0.07    
 SPQ-BRU −0.02 0.01 −0.12 −1.69 0.09    

Cog. Fx, cognitive functioning; PANAS–, Positive and Negative Affect Schedule, Negative subscale; SPQ-BRU, Schizotypal Personality Questionnaire-Brief Revised (Updated); DERS, Difficulties in Emotion Regulation Scale; IPANAT-NA, Implicit Positive and Negative Affect Test, Negative Affect subscale; SSFT, Strange Stories Film Task; B, unstandardized beta; B (SE), standard error of unstandardized beta; β, standardized beta; t, t test statistic; p, significance.

*p < 0.05, ***p < 0.001.

aF(4,196) = 9.51, p < 0.001, Adj. R2 = 0.15.

bF(5,195) = 8.25, p < 0.001, Adj. R2 = 0.15.

Specific Schizotypy Dimensions and ToM

It was hypothesized that there would be a significant, negative effect of specific schizotypy dimensions on ToM task performance, as measured by the FHA’s two subscales (FHA-Cat and FHA-Feel) and the SSFT.

Frith-Happé Animations (FHA). A hierarchical multiple regression analysis was conducted with age, Cog. Fx, and gender as covariates with positive schizotypy, negative schizotypy, and disorganized schizotypy entered as predictor variables and ToM, as measured by the FHA-Cat subscale as the outcome variable (Table 6). The overall regression was significant, F(6,194) = 8.98, p < 0.001, Adj. R2 = 0.18. There was a significant effect of negative schizotypy on ToM, b = −0.11, t(194) = −2.7, p = 0.008. There were no significant relationships found between positive schizotypy and ToM or disorganized schizotypy and ToM. The results indicate that negative schizotypy, but not positive or disorganized schizotypy, impacts ToM accuracy, as measured by the FHA-Cat. Another hierarchical multiple regression analysis was conducted with age, Cog. Fx, and gender as covariates with positive schizotypy, negative schizotypy, and disorganized schizotypy entered as predictor variables and ToM, as measured by the FHA-Feel as the outcome variable (Table 7). The overall regression was significant, F(6,194) = 8.98, p < 0.001, Adj. R2 = 0.19. There was a significant effect of negative schizotypy on ToM, b = −0.09, t(194) = −2.45, p = 0.02. The results indicate that negative schizotypy, but not positive or disorganized schizotypy, impacts ToM accuracy, as measured by the FHA-Feel subscale.

Table 6.

Hierarchical multiple regression results with specific schizotypy dimensions as independent variables and FHA-Cat as dependent variable

BB (SE)βtp valueRR2R2
Step 1a      0.43 0.18 0.18 
 (Constant) 8.32 1.59  5.22 <0.001***    
 Age −0.12 0.07 −0.11 −1.74 0.08    
 Cog. Fx 0.30 0.05 0.37 5.73 <0.001***    
 Gender −0.63 0.29 −0.14 −2.15 0.03*    
Step 2b      0.47 0.22 0.03 
 (Constant) 8.27 1.64  5.04 <0.001    
 Age −0.10 0.07 −0.09 −1.38 0.17    
 Cog. Fx 0.29 0.05 0.36 5.54 <0.001    
 Gender −0.62 0.29 −0.14 −2.12 0.04*    
 MSS-P 0.03 0.03 0.07 0.89 0.38    
 MSS-N −0.11 0.04 −0.20 −2.70 0.008***    
 MSS-D 0.01 0.03 0.01 0.15 0.88    
BB (SE)βtp valueRR2R2
Step 1a      0.43 0.18 0.18 
 (Constant) 8.32 1.59  5.22 <0.001***    
 Age −0.12 0.07 −0.11 −1.74 0.08    
 Cog. Fx 0.30 0.05 0.37 5.73 <0.001***    
 Gender −0.63 0.29 −0.14 −2.15 0.03*    
Step 2b      0.47 0.22 0.03 
 (Constant) 8.27 1.64  5.04 <0.001    
 Age −0.10 0.07 −0.09 −1.38 0.17    
 Cog. Fx 0.29 0.05 0.36 5.54 <0.001    
 Gender −0.62 0.29 −0.14 −2.12 0.04*    
 MSS-P 0.03 0.03 0.07 0.89 0.38    
 MSS-N −0.11 0.04 −0.20 −2.70 0.008***    
 MSS-D 0.01 0.03 0.01 0.15 0.88    

Cog. Fx, cognitive functioning; MSS-P, Multidimensional Schizotypy Scale, Positive Schizotypy subscale; MSS-N, Multidimensional Schizotypy Scale, Negative Schizotypy subscale; MSS-D, Multidimensional Schizotypy Scale, Disorganized Schizotypy subscale; FHA-Cat, Frith-Happé animations, Categorization subscale; B, unstandardized beta; B (SE), standard error of unstandardized beta; β, standardized beta; t, t test statistic; p, significance.

*p < 0.05, ***p < 0.001.

aF(3,197) = 14.82, p < 0.001, Adj. R2 = 0.17.

bF(6,194) = 8.98, p < 0.001, Adj. R2 = 0.19.

Table 7.

Hierarchical multiple regression results for FHA-Feel as dependent variable

BB (SE)βtp valueRR2R2
Step 1a      0.44 0.19 0.19 
 (Constant) 4.40 1.37  3.22 <0.001***    
 Age −0.13 0.06 −0.13 −2.07 0.04*    
 Cog. Fx 0.26 0.05 0.37 5.71 <0.001***    
 Gender −0.60 0.25 −0.15 −2.40 0.02*    
Step 2b      0.47 0.22 0.02 
 (Constant) 4.22 1.41  2.98 0.003**    
 Age −0.10 0.06 −0.10 −1.59 0.11    
 Cog. Fx 0.25 0.05 0.36 5.56 <0.001***    
 Gender −0.63 0.25 −0.16 −2.50 0.01*    
 MSS-P 0.00 0.03 0.01 0.08 0.94    
 MSS-N −0.09 0.04 −0.18 −2.45 0.02*    
 MSS-D 0.03 0.03 0.08 0.97 0.33    
BB (SE)βtp valueRR2R2
Step 1a      0.44 0.19 0.19 
 (Constant) 4.40 1.37  3.22 <0.001***    
 Age −0.13 0.06 −0.13 −2.07 0.04*    
 Cog. Fx 0.26 0.05 0.37 5.71 <0.001***    
 Gender −0.60 0.25 −0.15 −2.40 0.02*    
Step 2b      0.47 0.22 0.02 
 (Constant) 4.22 1.41  2.98 0.003**    
 Age −0.10 0.06 −0.10 −1.59 0.11    
 Cog. Fx 0.25 0.05 0.36 5.56 <0.001***    
 Gender −0.63 0.25 −0.16 −2.50 0.01*    
 MSS-P 0.00 0.03 0.01 0.08 0.94    
 MSS-N −0.09 0.04 −0.18 −2.45 0.02*    
 MSS-D 0.03 0.03 0.08 0.97 0.33    

Cog. Fx, cognitive functioning; SPQ, Schizotypal Personality Questionnaire-Brief Revised (Updated); FHA-Cat, Frith-Happé animations, Categorization subscale; MSS-P, Multidimensional Schizotypy Scale, Positive Schizotypy subscale; MSS-N, Multidimensional Schizotypy Scale, Negative Schizotypy subscale; MSS-D, Multidimensional Schizotypy Scale, Disorganized Schizotypy subscale; B, unstandardized beta; B (SE), standard error of unstandardized beta; β, standardized beta; t, t test statistic; p, significance.

*p < 0.05, **p < 0.01, ***p < 0.001.

aF(3,197) = 15.70, p < 0.001, Adj. R2 = 0.18.

bF(6,194) = 8.98 p < 0.001, Adj. R2 = 0.19.

Strange Stories Film Task (SSFT). A hierarchical multiple regression analysis was conducted with age, Cog. Fx, explicit negative affect, and ethnicity as covariates with positive schizotypy, negative schizotypy, and disorganized schizotypy entered as predictor variables and ToM, as measured by the SSFT as the outcome variable (Table 8). The overall regression was significant, F(7,193) = 7.86, p < 0.001, R2 = 0.22. There was a significant effect of negative schizotypy on ToM, b = −0.12, t(193) = −3.22 p = 0.001. However, there were no significant relationships found between positive schizotypy and ToM or disorganized schizotypy and ToM. The results indicate that negative schizotypy, but not positive or disorganized schizotypy, impacts ToM accuracy, as measured by the SSFT.

Table 8.

Hierarchical multiple regression results with specific schizotypy dimensions as independent variables and SSFT as dependent variable

BB (SE)βtp valueRR2R2
Step 1a      0.40 0.16 0.16 
 (Constant) 6.55 1.51  4.34 <0.001***    
 Age −0.16 0.06 −0.16 −2.45 0.02*    
 Cog. Fx 0.20 0.05 0.28 4.18 <0.001***    
 PANAS– −0.03 0.02 −0.09 −1.34 0.18    
 Ethnicity 0.71 0.36 0.13 1.98 0.05    
Step 2b      0.47 0.22 0.07 
 (Constant) 7.29 1.50  4.86 <0.001**    
 Age −0.14 0.06 −0.15 −2.22 0.03*    
 Cog. Fx 0.19 0.05 0.26 4.04 <0.001***    
 PANAS– 0.00 0.02 0.00 −0.06 0.95    
 Ethnicity 0.31 0.36 0.06 0.84 0.40    
 MSS-P −0.02 0.03 −0.05 −0.72 0.48    
 MSS-N −0.12 0.04 −0.24 −3.22 0.001**    
 MSS-D −0.01 0.03 −0.03 −0.39 0.70    
BB (SE)βtp valueRR2R2
Step 1a      0.40 0.16 0.16 
 (Constant) 6.55 1.51  4.34 <0.001***    
 Age −0.16 0.06 −0.16 −2.45 0.02*    
 Cog. Fx 0.20 0.05 0.28 4.18 <0.001***    
 PANAS– −0.03 0.02 −0.09 −1.34 0.18    
 Ethnicity 0.71 0.36 0.13 1.98 0.05    
Step 2b      0.47 0.22 0.07 
 (Constant) 7.29 1.50  4.86 <0.001**    
 Age −0.14 0.06 −0.15 −2.22 0.03*    
 Cog. Fx 0.19 0.05 0.26 4.04 <0.001***    
 PANAS– 0.00 0.02 0.00 −0.06 0.95    
 Ethnicity 0.31 0.36 0.06 0.84 0.40    
 MSS-P −0.02 0.03 −0.05 −0.72 0.48    
 MSS-N −0.12 0.04 −0.24 −3.22 0.001**    
 MSS-D −0.01 0.03 −0.03 −0.39 0.70    

Cog. Fx, cognitive functioning; PANAS–, Positive and Negative Affect Schedule, Negative subscale; MSS-P, Multidimensional Schizotypy Scale, Positive Schizotypy subscale; MSS-N, Multidimensional Schizotypy Scale, Negative Schizotypy subscale; MSS-D, Multidimensional Schizotypy Scale, Disorganized Schizotypy subscale; SSFT, Strange Stories Film Task; B, unstandardized beta; B (SE), standard error of unstandardized beta; β, standardized beta; t, t test statistic; p, significance.

*p < 0.05, **p < 0.01, ***p < 0.001.

aF(4,196) = 9.05 p < 0.001, Adj. R2 = 0.14.

bF(7,193) = 7.86, p < 0.001, Adj. R2 = 0.19.

Hypothesis Testing with EGD

Further analyses were conducted using an EGD approach in which cutoff scores on schizotypy measures were used to create low and high schizotypy groups to use for mean comparison tests. Mean comparison tests were conducted for each schizotypy variable as the grouping variable. Analyses were conducted with low and high groups formed out of the given schizotypy variable’s bottom 30% and top 30% scores. The 30% threshold was chosen arbitrarily as we could not find a standard threshold in our review of the schizotypy literature and there does not seem to be a standard threshold in studies that use EGD approach with dimension data [16]. For each schizotypy variable, a one-way ANOVA test was conducted as the most parsimonious analysis to determine whether there was a significant main effect of schizotypy on the three ToM scales.

Positive Schizotypy EGD

Low and high positive schizotypy groups were formed using cutoff scores in the following manner. The 30th percentile score (4) was used to create the low positive schizotypy group (M = 5.93, SD = 4.39, n = 61), and the 70th percentile score (11) was used to create the high positive schizotypy group (M = 9.88, SD = 5.88, n = 66). A one-way ANOVA was conducted to test whether there was a significant main effect of positive schizotypy on the three ToM scales, FHA-Cat, FHA-Feel, and SSFT. The main effect of positive schizotypy on SSFT was significant with a small effect, F(1, 125) = 7.38, p = 0.008, η2 = 0.03. The results indicate that the high positive schizotypy group (M = 4.51, SD = 2.38) performed significantly lower than the low positive schizotypy group (M = 5.66, SD = 2.35), which is consistent with the current study hypotheses. However, the main effect of positive schizotypy on FHA-Cat was not significant, F(1, 125) = 0.79, p = 0.38, η2 = 0.00, and the main effect of positive schizotypy on FHA-Feel was not significant, F(1, 125) = 0.61, p = 0.44, η2 = 0.00.

Negative Schizotypy EGD

Low and high negative schizotypy groups were formed using cutoff scores in the following manner. The 30th percentile score (2) was used to create the low negative schizotypy group (M = 3.80, SD = 4.08, n = 68), and the 70th percentile score (8) was used to create the high negative schizotypy group (M = 8.48, SD = 4.44, n = 68). A one-way ANOVA was conducted to test whether there was a significant main effect of negative schizotypy on the three ToM scales, FHA-Cat, FHA-Feel, and SSFT. The main effect of negative schizotypy on FHA-Cat was significant with a small effect, F(1, 134) = 9.08, p = 0.003, η2 = 0.11: the high negative schizotypy group (M = 5.97 SD = 2.67) performed significantly lower than the low negative schizotypy group (M = 7.30, SD = 2.50) on the FHA-Cat. The main effect of negative schizotypy on FHA-Feel was significant, F(1, 134) = 4.92, p = 0.03, η2 = 0.06: the high negative schizotypy group (M = 1.94, SD = 2.11) performed significantly lower than the low negative schizotypy group (M = 2.78, SD = 2.29) on the FHA-Feel. The main effect of negative schizotypy on SSFT was significant, F(1, 134) = 16.77, p < 0.001, η2 = 0.04: the high negative schizotypy group (M = 4.10, SD = 2.57) performed significantly lower than the low negative schizotypy group (M = 5.81, SD = 2.32) on the SSFT. These results are consistent with the current study hypotheses regarding the significant, negative relationship between the level of schizotypy and ToM accuracy.

Disorganized Schizotypy EGD

Low and high disorganized schizotypy groups were formed using cutoff scores in the following manner. The 30th percentile score (3) was to create a low disorganized schizotypy group (M = 1.16, SD = 1.07, n = 70), and the 70th percentile score (12) was used to create a high disorganized schizotypy group (M = 15.97, SD = 3.26, n = 62). A one-way ANOVA was conducted to test whether there was a significant main effect of disorganized schizotypy on the three ToM scales, FHA-Cat, FHA-Feel, and SSFT. The main effect of disorganized schizotypy on SSFT was significant with a small effect, F(1, 130) = 7.43, p = 0.007, η2 = 0.05; the high disorganized schizotypy group (M = 4.61, SD = 2.36) performed significantly lower than the low disorganized schizotypy group (M = 5.72, SD = 2.30) on the SSFT. However, the main effect of disorganized schizotypy on FHA-Cat was not significant, F(1, 130) = 0.14, p = 0.71, η2 = 0.00, and the main effect of disorganized schizotypy on FHA-Feel was not significant, F(1, 130) = 0.03, p = 0.87, η2 = 0.00. These results are in part consistent with the current study hypotheses regarding the significant, negative relationship between the level of schizotypy and ToM accuracy.

The current study sought to replicate results found in the schizophrenia literature that have found ToM impairments in clinical samples (for a meta-analysis, see [29]). Contrary to what was predicted, overall schizotypy was not found to be associated with ToM task performance. These null results are consistent with the subset of studies (e.g., [30, 31]) that have failed to find an association between schizotypy and ToM in nonclinical samples. However, the current study’s null results are inconsistent with other studies (e.g., [11, 12]) that have found significant, negative relationships between schizotypy and ToM. The discrepant findings in the literature suggest that potential sources of inconsistency might be methodological issues, such as inadequate attention to the multidimensional character of schizotypy itself, as well as the statistical methods used to examine ToM performance in schizotypy.

With the aim of accounting for the multidimensional nature of schizotypy, the current study evaluated whether there were significant, negative relationships between positive, negative, and disorganized schizotypy dimensions and ToM task performance. While a significant, negative relationship between negative schizotypy and ToM was found on both the FHA and SSFT, there was no significant relationship between positive schizotypy and ToM and disorganized schizotypy and ToM. This finding is consistent with the schizophrenia and schizotypy literature that view ToM deficits as fundamentally linked to the interpersonal deficits that characterize negative schizotypy [32]. Given that no significant relationship was found between the positive and disorganized dimensions and ToM in the multiple regression analysis, this finding suggests that the negative schizotypy dimension displays a uniquely pronounced impairment in ToM in nonclinical schizotypy. This finding can aid in early detection of psychotic illness as it suggests that ToM impairment is a potential risk factor for psychosis, which can inform treatment of individuals who present with elevated negative symptomatology.

Regarding statistical approaches, the current study sought to advance the evaluation of schizotypy by using both regression-based approach for hypothesis testing and an EGD paradigm that sorts participants into low and high schizotypy groups and employs mean comparison methods, such as ANOVA (for review and meta-analysis, see [33]). These two approaches are not mutually exclusive, and indeed they share a common mathematical foundation [34]. However, in studies of ToM in schizotypy, one consideration that influences choice of statistical method is theoretical views about the latent structure of schizotypy, i.e., whether schizotypy is fundamentally dimensional or taxonic in structure. The view of schizotypy as inherently dimensional lends itself to a regression-based approach, while the other dominant view of schizotypy as taxonic/categorical lends itself to a mean comparison approach. A common EGD procedure is to first screen hundreds of participants using a schizotypy scale and then sort participants into a low and high group according to a cutoff score, e.g., 2.5 SDs above the sample mean, or the top 10th percentile. In choosing the regression-based method for primary hypothesis testing, the current study was guided, in part, by sampling convenience and lack of resources to screen hundreds of participants, and by the theoretical view that schizotypy is a dimensional phenomenon and should therefore be studied continuously and not categorically. The current study explored how the EGD approach might provide a different perspective on the relationship between schizotypy and ToM. By comparing low and high schizotypy groups, a unique pattern emerged such that the high groups of all three specific schizotypy dimensions (positive, negative, and disorganized) performed lower than the corresponding low schizotypy groups on the SSFT ToM measure. In addition, the high negative schizotypy group also performed worse on the FHA-Cat and FHA-Feel.

The results of these analyses suggest several important implications. First, there is the methodological/statistical importance of the fact that the EGD approach yielded significant relationships between positive schizotypy and disorganized schizotypy and ToM, as measured by the SSFT. Since these relationships were not detected using the regression analyses, these results suggest the use of the EGD approach to detect significant effects that might be lost when the entire distribution of scores is entered into a regression model. When conducting a regression, it is possible that the middle range of scores that occur between the low and high scores “washes out” these effects in the regression analyses [16].

Second, the significant results found using the EGD approach, and not the regression-based approach, provide some support for the taxonic/categorical view of schizotypy. One way of explaining the differences in ToM performance between the low and high schizotypy groups is to say that these groups represent two qualitatively different categories or classes of individuals, e.g., non-schizotypes (i.e., the low schizotypy group) and schizotypes (i.e., the high schizotypy group). However, such a strong interpretation of the current study’s results does not follow merely because the EGD found significant effects. It is important to emphasize that a statistical model by itself, without multiple points of other converging evidence, should not determine the latent construct of a psychological construct, such as schizotypy [3]. A more conservative statement regarding the implications of the EGD analyses is that ToM deficits are detectable at the “extreme” high end of the schizotypy spectrum.

Third, there is the broader significance that this finding is consistent with the theoretical and empirical literature that has demonstrated a link between ToM impairments in schizophrenia and nonclinical schizotypy. The current study suggests that ToM is a deficit that is shared by all schizotypy dimensions, given that all three specific schizotypy dimensions demonstrated weaker performance on the SSFT compared to the corresponding low schizotypy groups. However, it also suggests that the ToM impairment is expressed differentially across schizotypy dimensions, as the high negative schizotypy group also demonstrated weaker performance on the FHA ToM scales.

The results of the current study should be considered in the context of several limitations. One limitation is related to the generalizability of the current study’s findings. While the ethnic diversity of the sample (75.6% non-white) might in another context be a strength, it is difficult to compare the findings with other schizotypy research that use overwhelmingly white samples. Another important limitation of the current study is the failure to screen for current non-schizotypic psychopathology, and so other symptomatology might be an additional variable that can influence performance on ToM tasks. Several limitations relate to the ToM measures used in the current study. Regarding ToM measures, the current study is consistent with reviews of the ToM measurement literature that stress the difficulties involved in assessing ToM [33]. The current study sought to compare participants’ performance on two very different ToM tasks, one that consisted of minimal culturally dependent content and verbal comprehension skills (FHA) and another task that was laden with cultural references and required further language processing skills (SSFT). While the SSFT is an ecologically valid video-based measure of ToM that is based on the widely used written original Strange Stories ToM task, there are some aspects of the SSFT that might limit its application in some participant pools. First, the SSFT depicts only white characters. Given the high proportion of non-white participants, it is possible that the homogenous racial composition of the film characters might have impacted participant responses. In addition, the actors in the SSFT film clips spoke with British accents that might make it difficult for North American participants to understand, particularly for participants who speak English as a second language. As such, within the current study sample, it is possible, e.g., that an incorrect response on an SSFT item was due to difficulty understanding what a character was saying rather than a failure of ToM.

The present study’s findings indicate several directions for future research. First, given the strong link between negative schizotypy and ToM impairment found in the present study, further research should explore whether specific features of negative schizotypy might explain this link. For example, social anhedonia, the reduced capacity to experience pleasure in social settings, might contribute to ToM deficits due to diminished practice of social cognitive skills [35], and so a future study might include a measure that specifically taps this construct (e.g., [36]). In addition, several suggestions emerge out of the limitations of the current study concerning sample selection and use of between-group designs that compare low and high schizotypy groups along with a non-schizotypic psychopathology group to identify ToM impairment in specific etiology for schizophrenia. For example, given the current study’s significant findings using the EGD approach that were not matched in the regression-based approach, future studies should consider using an EGD approach for studying schizotypy in ToM. We used a cutoff threshold of the bottom 30% and the top 30% to define low and high schizotypy groups. However, given the lack of a standard cutoff threshold in the schizotypy literature, future studies might explore how different thresholds might impact statistical power and significance testing. In addition, a recommendation to future studies would be the inclusion of a non-schizotypic psychopathology “control” group. While the current study implemented an EGD whereby participants were sorted into low and high schizotypy groups, it did not implement a plan for generating a third group that would tap higher levels of non-schizotypic psychopathology. To add further specificity in schizotypy research, researchers have encouraged the use of not only “healthy” or “low schizotypy” comparison/control groups but also a general (non-schizotypic) psychopathology control group to help bolster a study’s claims about specific etiology in schizotypy. Without the use of a general psychopathology group, it is not clear whether the findings of schizotypy study reflect something specific about schizotypy or about general psychopathology or general psychological distress.

This study protocol was reviewed and approved by the Institutional Review Board (IRB) at Long Island University, Approval No. 22/02-021. Participants indicated their consent via online consent form that was approved by the IRB at Long Island University, Approval No. 22/02-021. Participants were presented with the informed consent form with IRB approval stamp, and they indicated their consent to participate by clicking a button (“I agree to participate.”) at the bottom of the informed consent form.

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funder.

R.A.N.G. contributed to the conceptualization, methodological design, data collection, and analysis, and wrote the original draft of this paper. N.P. and K.B.M. supervised the dissertation research project, contributed to the conceptualization of the study and interpretation of the results, and assisted with the revision and editing of the manuscript. M.J.M. and D.K. contributed to editing and revision of the manuscript. All authors approved the final version.

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants but are available from the corresponding author [R.A.N.G.] upon reasonable request.

1.
Kwapil
TR
,
Barrantes-Vidal
N
.
Schizotypy: looking back and moving forward
.
Schizophr Bull
.
2015
;
41
(
Suppl 2
):
S366
73
.
2.
Lenzenweger
MF
.
Schizotypy, schizotypic psychopathology and schizophrenia
.
World Psychiatr
.
2018
;
17
(
1
):
25
6
.
3.
Lenzenweger
MF
.
Schizotypy and schizophrenia: the view from experimental psychopathology
.
Guilford Press
;
2011
.
4.
Meehl
PE
.
Schizotaxia, schizotypy, schizophrenia
.
Am Psychol
.
1962
;
17
(
12
):
827
38
.
5.
Chapman
LJ
,
Chapman
JP
,
Kwapil
TR
,
Eckblad
M
,
Zinser
MC
.
Putatively psychosis-prone subjects 10 years later
.
J Abnorm Psychol
.
1994
;
103
(
2
):
171
183
.
6.
Cannon
TD
.
Clinical and genetic high-risk strategies in understanding vulnerability to psychosis
.
Schizophr Res
.
2005
;
79
(
1
).
7.
Nelson
MT
,
Seal
ML
,
Pantelis
C
,
Phillips
LJ
.
Evidence of a dimensional relationship between schizotypy and schizophrenia: a systematic review
.
Neurosci Biobehav Rev
.
2013
;
37
(
3
):
317
27
.
8.
Fonagy
P
,
Target
M
.
Attachment and reflective function: their role in self-organization
.
Dev Psychopathol
.
1997
;
9
(
4
):
679
700
.
9.
Debbané
M
,
Eliez
S
,
Badoud
D
,
Conus
P
,
Fluckiger
R
,
Schultze-Lutter
F
.
Developing psychosis and its risk states through the lens of schizotypy
.
Schizophr Bull
.
2015
;
41
(
suppl 2
):
S396
407
.
10.
Premack
D
,
Woodruff
G
.
Does the chimpanzee have a theory of mind
.
Behav Brain Sci
.
1978
;
1
(
4
):
515
26
.
11.
Henry
JD
,
Green
MJ
,
Restuccia
C
,
de Lucia
A
,
Rendell
PG
,
McDonald
S
, et al
.
Emotion dysregulation and schizotypy
.
Psychiatry Res
.
2009
;
166
(
2–3
):
116
24
.
12.
Langdon
R
,
Coltheart
M
.
Mentalising, schizotypy, and schizophrenia
.
Cognition
.
1999
;
71
(
1
):
43
71
.
13.
Fyfe
S
,
Williams
C
,
Mason
O
,
Pickup
G
.
Apophenia, theory of mind and schizotypy: perceiving meaning and intentionality in randomness
.
Cortex
.
2008
;
44
(
10
):
1316
25
.
14.
Pickup
G
.
Theory of mind and its relation to schizotypy
.
Cogn Neuropsychiatry
.
2006
;
11
(
2
):
117
92
.
15.
Claridge
G
.
Single indicator of risk for schizophrenia: probable fact or likely myth
.
Schizophr Bull
.
1994
;
20
(
1
):
151
68
.
16.
Fisher
JE
,
Guha
A
,
Heller
W
,
Miller
GA
.
Extreme-groups designs in studies of dimensional phenomena: advantages, caveats, and recommendations
.
J Abnorm Psychol
.
2020
;
129
(
1
):
14
20
.
17.
Kwapil
TR
,
Gross
GM
,
Silvia
PJ
,
Raulin
ML
,
Barrantes-Vidal
N
.
Development and psychometric properties of the Multidimensional Schizotypy Scale: a new measure for assessing positive, negative, and disorganized schizotypy
.
Schizophr Res
.
2018
;
193
:
209
17
.
18.
Murray
K
,
Johnston
K
,
Cunnane
H
,
Kerr
C
,
Spain
D
,
Gillan
N
, et al
.
A new test of advanced theory of mind: the “Strange Stories Film Task” captures social processing differences in adults with autism spectrum disorders
.
Autism Res
.
2017
;
10
(
6
):
1120
32
.
19.
Livingston
LA
,
Shah
P
,
White
SJ
,
Happé
F
.
Further developing the Frith–Happé animations: a quicker, more objective, and web-based test of theory of mind for autistic and neurotypical adults
.
Autism Res
.
2021
;
14
(
9
):
1905
12
.
20.
White
SJ
,
Coniston
D
,
Rogers
R
,
Frith
U
.
Developing the Frith-Happé animations: a quick and objective test of Theory of Mind for adults with autism
.
Autism Res
.
2011
;
4
(
2
):
149
54
.
21.
Happé
FG
.
An advanced test of theory of mind: understanding of story characters’ thoughts and feelings by able autistic, mentally handicapped, and normal children and adults
.
J Autism Dev Disord
.
1994
;
24
(
2
):
129
54
.
22.
Davidson
CA
,
Hoffman
L
,
Spaulding
WD
.
Schizotypal personality questionnaire–brief revised (updated): an update of norms, factor structure, and item content in a large non-clinical young adult sample
.
Psychiatry Res
.
2016
;
238
:
345
55
.
23.
Martinez
G
,
Mosconi
E
,
Daban-Huard
C
,
Parellada
M
,
Fananas
L
,
Gaillard
R
, et al
.
“A circle and a triangle dancing together”: alteration of social cognition in schizophrenia compared to autism spectrum disorders
.
Schizophr Res
.
2019
;
210
:
94
100
.
24.
Condon
DM
,
Revelle
W
.
The international cognitive ability resource: development and initial validation of a public-domain measure
.
Intell
.
2014
;
43
:
52
64
.
25.
Watson
D
,
Clark
LA
,
Tellegen
A
.
Development and validation of brief measures of positive and negative affect: the PANAS scales
.
J Pers Soc Psychol
.
1988
;
54
(
6
):
1063
70
.
26.
Jackson
DN
. Personality research form. Incorporated:
Research Psychologists Press
;
1965
.
27.
IBM Corp
.
IBM SPSS statistics for mac (version 28.0) [Computer software]
.
IBM Corp
.
2021
.
28.
Faul
F
,
Erdfelder
E
,
Lang
AG
,
Buchner
A
.
G* Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences
.
Behav Res Methods
.
2007
;
39
(
2
):
175
91
.
29.
Savla
GN
,
Vella
L
,
Armstrong
CC
,
Penn
DL
,
Twamley
EW
.
Deficits in domains of social cognition in schizophrenia: a meta-analysis of the empirical evidence
.
Schizophr Bull
.
2013
;
39
(
5
):
979
92
.
30.
Fernyhough
C
,
Jones
SR
,
Whittle
C
,
Waterhouse
J
,
Bentall
RP
.
Theory of mind, schizotypy, and persecutory ideation in young adults
.
Cogn Neuropsychiatry
.
2008
;
13
(
3
):
233
49
.
31.
Jahshan
CS
,
Sergi
MJ
.
Theory of mind, neurocognition, and functional status in schizotypy
.
Schizophr Res
.
2007
;
89
(
1–3
):
278
86
.
32.
Frith
CD
.
The cognitive neuropsychology of schizophrenia
.
Hove, East Sussex
:
Psychology Press
;
2015
.
33.
Eddy
CM
.
What do you have in mind? Measures to assess mental state reasoning in neuropsychiatric populations
.
Front Psychiatry
.
2019
;
10
:
425
.
34.
Cohen
J
,
Cohen
P
,
West
SG
,
Aiken
LS
.
Applied multiple regression/correlation analysis for the behavioral sciences
.
Routledge
;
2013
.
35.
Dodell-Feder
D
,
Tully
LM
,
Lincoln
SH
,
Hooker
CI
.
The neural basis of theory of mind and its relationship to social functioning and social anhedonia in individuals with schizophrenia
.
NeuroImage Clin
.
2014
;
4
:
154
63
.
36.
Eckblad
ML
,
Chapman
LJ
,
Chapman
JP
,
Mishlove
M
. The revised social anhedonia scale. Unpublished test;
1982
.
37.
Brent
BK
,
Fonagy
P
.
A mentalization-based treatment approach to disturbances of social understanding in schizophrenia
. In:
Lysaker
P
,
Dimaggio
G
,
Brüne
M
, editors.
Social cognition and metacognition in schizophrenia: psychopathology and treatment approaches
.
Elsevier
;
2014
. p.
245
59
.
38.
Brune
M
.
“Theory of mind” in schizophrenia: a review of the literature
.
Schizophr Bull
.
2005
;
31
(
1
):
21
42
.
39.
Abell
F
,
Happe
F
,
Frith
U
.
Do triangles play tricks? Attribution of mental states to animated shapes in normal and abnormal development
.
Cogn Dev
.
2000
;
15
(
1
):
1
16
.
40.
Bora
E
.
Theory of mind and schizotypy: a meta-analysis
.
Schizophr Res
.
2020
;
222
:
97
103
.