Cognitive failures and the role of emotion in dimensional schizotypy: A replication and extension

Abstract

Introduction:

Cognitive failures are commonplace within the general population but may be particularly heightened in those with higher levels of schizotypy. This is especially salient in the context of enduring trait and momentary state negative emotion which often contributes to increases in daily impairments, leading to a more debilitating and distracted life. Particularly, individuals with elevated levels of schizotypy may be more likely to experience cognitive failures, especially in the presence of negative trait emotion such as depression, anxiety, and stress. However, little is known about the influence of state emotion and the distinct roles that state and trait emotion may have with cognitive failures and schizotypy.

Methods:

To replicate and extend previous findings, 306 (58% males) undergraduate students, aged 18–50 (M=19.343; SD=2.493) completed self-report measures of cognitive failures, trait and state emotion, and schizotypy. Mediation and moderation analyses were performed in SPSS to examine the potential effects of trait and state emotion on the relationship between schizotypy and cognitive failures.

Results:

Consistent with previous findings, mood symptomology, in addition to negative affect, mediated cognitive failures in those with higher levels of schizotypy. However, in our sample, positive affect did not appear to buffer against cognitive failures.

Discussion/Conclusion:

The findings of the present study suggest there may be a nuanced relationship between both negative trait and state emotions on the relationships between cognitive failures and schizotypy. Understanding the interaction of enduring versus momentary emotion on cognition as they relate to an elevated risk for developing schizophrenia-spectrum phenomena may be a point for earlier and more targeted interventions.

Introduction

Cognitive failures, such as forgetting appointments and misplacing keys, are normative daily occurrences. The frequency of cognitive failures increases with negative emotion, and there is some evidence to suggest that this relationship is more impactful in those at risk for schizophrenia-spectrum symptoms [1, 2]. Negative emotion may potentially exacerbate cognitive deficits such as working memory and informational processing impairments that further contribute to the emergence of symptoms along the schizophrenia-spectrum [3–5]. Similarly, patterns of negative trait emotion (e.g., depression, anxiety, stress) appear to influence cognitive failures within the schizophrenia spectrum [1]. However, there is evidence suggesting that acute fluctuations in psychosis experienced by individuals with schizophrenia may be more influenced by momentary state emotion. For example, individuals who experienced negative state mood reported increases in momentary psychosis while other individuals in response to elevated positive state mood experienced lower expressions of momentary psychosis [6]. While enduring patterns of negative emotions may contribute to the risk for psychosis (i.e., schizotypy) and cognitive failures, little is understood about how in-the-moment positive (e.g., excited, enthusiastic, proud) and negative (e.g., distressed, irritable, ashamed) state emotion influence cognitive failures in those with schizotypy. Therefore, the present study aimed to replicate, as presented by Carrigan and Barkus [1], the influence of emotion on schizotypy and cognitive failures. We expanded these findings to explore in-the-moment state and positive emotion as they pertain to cognitive failures and levels of schizotypy.

Cognitive failures may be influenced by in-the-moment experiences and are a highly contextual aspect of cognitive capacity. This contrasts with cognitive deficits which are considered more stable representations of one’s functional performance based on objective cognitive assessments [7]. While cognitive deficits and failures together can provide a richer understanding of one’s cognitive functioning, individually, deficits and failures reflect distinct phenomena. For instance, objective assessments for cognitive deficits typically occur in highly controlled testing environments which may not reflect real-life circumstances. Cognitive failures, a more subjective measure of cognitive processing, may more so reflect day-to-day experiences in an uncontrolled environment [7]. Thus, cognitive failures may represent daily functional impairments as opposed to impairments of objective cognitive performance [8, 9]. As such, frequent cognitive failures can heighten distress, especially when they interfere with one’s career, school, and social life [7]. The disruptions and distress associated with cognitive failures may be especially salient in the context of the schizophrenia spectrum as deficits in cognitive functioning [10] and mood symptoms may influence impairments in everyday functioning [3]. For example, cognition, mood, and negative symptoms may be amplified by the presence of subjective experiences of cognitive failures [11]. The presence of these failures may reflect impairments and inefficiency in utilizing cognitive resources [12].

According to cognitive load theory, cognition can be viewed as a limited resource where increased demands are associated with decreased cognitive functioning, such as poor working memory and preprocessing of information [13, 14]. As cognitive failures may reflect a poor application of cognitive resources [12], individuals with higher levels of schizotypy may be more acutely affected by cognitive-resource demands [14, 15]. Thus, these individuals could be more susceptible to cognitive failures, especially in times of heightened emotion [1]. Specifically, it appears that cognitive failures form distinct relationships with certain subscales of schizotypy and depressive symptoms. For example, during higher cognitive load tasks, individuals with higher levels of negative schizotypy (i.e., deficits in interpersonal interactions) tend to exhibit more impoverished and poorer speech patterns than during lighter cognitive load tasks [16]. This pattern may be influenced by emotion as prior research suggests that cognitive failures emerge before the presence of negative schizotypy and depressive-like symptoms, further supporting a relationship between negative emotion in the context of schizotypy and cognitive failures [11].

Emotional impairments have been observed across the schizophrenia spectrum from experiences in at-risk individuals to clinical presentations of symptoms [17] and, as such, have the potential to exacerbate symptomology [18]. However, there have been conflicting definitions of emotion within the literature conflating and often interchanging trait emotion, which refers to enduring and longer lasting mood presentation, and state emotion, which refers to the in-the-moment experiences of positive and negative affect [19]. Distinguishing between these two concepts may provide a larger context for the role of emotion and cognition in the schizophrenia spectrum. There is some evidence that suggests the interaction of trait emotion, particularly negative trait emotion, and cognitive processes contribute to the development of positive symptoms, with mood symptoms (e.g., depression and anxiety) typically preceding hallucinations and delusions [20]. However, the influence of trait emotions on cognitive processes may not fully reflect how individuals interact with their environment. Examining the subtle fluctuations in an individual’s experience via state emotions in reaction to their environment may allow for further understanding of the risk for developing psychosis [21]. For instance, there appears to be a reciprocal relationship between negative and positive state emotion as they relate to momentary experiences of psychosis; positive state emotion tends to decrease momentarily hallucinatory experiences while negative state emotions appear to increase these experiences [6]. Exploring the nuances between state and trait emotion may thus present important distinctions necessary to disentangle their contributions to cognitive failures in schizotypy.

As the link between cognitive failures and negative trait emotion has been established in schizotypy [1], other distinct patterns of trait and state emotion have emerged across the schizophrenia spectrum. In particular, individuals with schizophrenia are more likely to experience lower instances of positive and higher rates of negative trait emotion [22]. While this pattern of reduced positive and heightened negative emotion is broadly paralleled in schizotypy [17], the distinction of state emotion may be important in relation to the subclinical dimensions (i.e., positive, negative, and disorganized). For example, although state emotion in negative schizotypy shares similar patterns as trait emotion in schizophrenia, there is evidence for higher negative affect, but not decreased positive affect in positive schizotypy (i.e., abnormal perceptual experiences) [23]. Furthermore, increased levels of negative schizotypy have been linked to impairments in experiencing, processing, and expressing emotions, while those with elevated disorganized schizotypy (i.e., poor coordination of thoughts) may have difficulty in regulating cognitions and emotion [24]. Similarly, impairments in emotion processing (e.g., attention to emotion information, emotion recognition) in those with higher levels of positive schizotypy may further influence cognitive and perceptual experiences [25]. Emotional dysfunction may be compounded by the presence of both positive and negative schizotypy, with negative schizotypy being related to increased impairments [26]. In particular, negative affect is viewed as a prominent etiological factor that likely coexists with additional biopsychosocial factors to influence the relationship of schizotypy and other constructs [27, 23, 1], including lower levels of life satisfaction in those with higher negative, disorganized, and global schizotypy [27].

The current study sought to replicate previous trait-related mood findings, while further examining the influence of state-related emotion on cognitive failures. First, we attempted to replicate research conducted by Carrigan & Barkus [1] by confirming the mediation of negative trait emotion on the relationship of cognitive failures across global, positive, negative, and disorganized schizotypy. Similar to these previous findings, we did not anticipate negative trait emotion to moderate this relationship. Second, as there is little literature exploring the role of state emotion within the relationship of cognitive failures and schizotypy, we extended previous findings to include positive and negative affect. Consistent with Carrigan and Barkus [1], we examined schizotypy as a dimensional and continuous construct. We predicted that our findings relating to state emotion would parallel previous trait emotion [1] with negative affect acting as a mediator between positive, negative, and disorganized schizotypy and cognitive failures. As an exploratory aim, we examined the role of positive affect on the relationship between global and subdomain schizotypy scores and cognitive failures, predicting a potential protective influence on cognitive failures. Understanding trait and state emotions as key etiological factors of the schizophrenia-spectrum as they relate to cognitive failures may therefore provide insights into more targeted early interventions.

Materials and Methods

Participants

Participants were recruited as part of their introductory psychology course. Individuals satisfied their course requirement by participation in this study or via alternative assignments (e.g., a research review). We collected data from 358 undergraduate students over the age of 18 at a northeastern university. Participants that had incomplete responses (n=8) or endorsed a contradictory value on one or both infrequency items (n=44) based on items developed by Chapman & Chapman [28] were omitted from analyses. As a result, fifty-two participants, who shared similar demographic patterns to our final sample, were excluded from the final analyses. Thus, our final sample consisted of 306 participants (males=178, females=127, other gender spectrum=1) with ages ranging from 18–50 years old (Mean=19.343, SD=2.493). The majority of our sample identified as Caucasian/White (69.3%), followed by African American/Black (11.4%), Hispanic/Latino (8.8%), multiple ethnicity endorsements (5.6%), Asian/Pacific Islander (3.6%), and other ethnicities (1.3%).

Procedures

The current study consists of a subset of data collected as part of a larger study that took approximately 90 minutes. All study-related activities were approved by a local institutional review board. Participants were recruited through an online study participation website where they were provided with initial information regarding the nature of the study. Participants selected a time for a one-time, face-to-face meeting with a trained research assistant. Before participation in the study, individuals provided informed consent. Data for the current study was collected using online questionnaires in a quiet and private research space. Due to the longer length of the protocol, infrequency items were added throughout the measures. Similarly, we counterbalanced three different orders of measures and tasks.

Measures

Schizotypal Personality Questionnaire-Brief Revised.

The SPQ-BR [29] measured levels of schizotypy across three symptom domains similar to those found in schizophrenia-spectrum conditions: positive, negative, and disorganized. Participants respond to each statement with Likert scale responses ranging from 1 (“Strongly Disagree”) to 5 (“Strongly Agree”) across 32 items. Positive schizotypy (i.e., odd perceptual experiences) items included “Do you ever suddenly feel distracted by distant sounds that you are not normally aware of?“ Negative schizotypy (i.e., social withdrawal and anhedonia) items included “I feel I have to be on my guard even with friends.” Disorganized schizotypy (i.e., odd thoughts) items included “I am an odd, unusual person.” This measure demonstrated good internal consistency with the current sample’s Cronbach’s alphas being α=.857 for negative schizotypy, α=.860 for positive schizotypy, α=.875 for disorganized schizotypy, and α=.922 for global schizotypy.

Extended Cognitive Failures Questionnaire.

The ECFQ [30] measured levels of self-reported cognitive failures and is based on the Cognitive Failures Questionnaire [31]. Self-report assessments of cognitive failures have been considered to be trait-like measurements of the tendency to experience cognitive failures and appear to remain stable overtime [31, 32] with good test-retest reliability [32].There is also some evidence to suggest that these trait-like experiences are influenced by state/contextual factors [7]. For the purpose of the present study, the ECFQ is considered to be a measure of trait-like experiences. The ECFQ consists of 35 items that involve everyday memory impairments and absent-mindedness. Items included questions such as, “Do you find you forget appointments?” and “Do you find you forget whether you’ve turned off a light or a fire or locked the door?” Participants rated their agreement to items on a 5-point Likert scale with 0 (“Never”) and 4 (“Very often”). Cronbach’s alpha revealed that this measure in the current study has excellent internal consistency (α=.922).

Positive and Negative Affect Schedule.

The PANAS [33] measured self-reported negative and positive state emotion (i.e., affect). This measure contained 20 items with 10 positively valenced and 10 negatively valenced emotion words. Participants were asked to rate each word with the extent that emotion applies to them in the present moment, right now from a Likert scale of 1 (“Very slightly or not at all”) to 5 (“Extremely”). Items included “Interested” and “Distressed.” Internal consistency was excellent for both subscales with positively valenced items having a Cronbach’s alpha of α=.913 and α=.896 for negatively valenced items in the current study.

Depression, Anxiety, Stress Scales.

Trait emotion was measured by the DASS [34]. A total score comprising three subscales (depression, anxiety, and stress) was used to represent overall enduring mood symptomatology. Participants rated statements as they apply to them over the past week from 0 (“Did not apply to me at all”) to 3 (“Applied to me very much, or most of the time”). Items included “I couldn’t seem to experience any positive feeling at all.” and “I was worried about situations in which I might panic and make a fool of myself.” In the current sample, Cronbach’s alpha for the total score was α=.961.

Statistical Analysis

We examined distributions across all variables for normality and evaluated skewness and kurtosis. Any value greater than ±2 was considered skewed or kurtote. Ethnicity and gender differences were assessed across all variables. ANOVA and individual samples t-tests were used for normative (parametric) data, and the Mann-Whitney U and Kruskal-Wallis tests for non-normally distributed data. When assessing for gender differences, there was one individual who endorsed other gender spectrum as their gender identity (shown in Table 1). As a result, the participant was not included in gender differences analyses but was included in regression and mediation analyses.

Table 1:

Means and correlations for variables of interest

	Mean (SD)	1	2	3	4	5	6	7	8
1. ECFQ	58.111 (17.386)	1
2. Global Schizotypy	48.726 (19.496)	.651**	1
3. Pos. Schizotypy	16.258 (9.000)	.529**	.851**	1
4. Neg. Schizotypy	17.026 (7.757)	.462**	.791**	.445**	1
5. Dis. Schizotypy	15.441 (6.738)	.644**	.846**	.616**	.542**	1
6. DASS total	21.023 (19.356)	.477**	.603**	.428**	.540**	.563**	1
7. Positive Affect	26.735 (7.724)	−.142*	−.160**	−.020	−.292**	−.100	−.249**	1
8. Negative Affect	17.026 (6.914)	.345**	.615**	.506**	.414**	.303**	.454**	.242**	1

Note. Spearman correlations presented in italics, ECFQ=Extended Cognitive Failures Questionnaire total score, Pos. Schizotypy=Positive schizotypy, Neg. Schizotypy=Negative schizotypy, Dis. Schizotypy=disorganized schizotypy, DASS total=depression, anxiety, stress.

^*denotes p<0.05;

^**denotes p<0.001.

To compare the relationships between variables in the current study to prior research [1], Pearson and Spearman correlations were performed. Furthermore, utilizing the Baron and Kenny Method, we explored the potential moderation of state (PA and NA scores as defined by the PANAS) and trait (DASS total scores) emotion between global and subscales scores of the SPQ-BR and the ECFQ. Scores for all variables were centered to the mean and entered in a hierarchical regression analysis following predefined metrics [35]. Additionally, gender was controlled for in the first step of all moderation models. Decomposition of interactions were performed using methods as described by Sibley [36].

We assessed the potential mediation of state and trait-like variables, as defined by the PANAS and DASS, on the relationship between the SPQ-BR and ECFQ. Measure scores were not mean-centered as there is some evidence to suggest that it may not be necessary and reduce multicollinearity [37]. Using the SPSS macro, PROCESS v3.4 [37], with 5,000 bootstrapped samples at a 95% confidence interval (CI) as defined by Hayes [37], we performed hierarchical regression analyses of the total, direct, and indirect effects. Additionally, the Baron and Kenny Method [35] was used to confirm mediation for the relationship between the total score on the SPQ-BR, the NA subscale on the PANAS, and total ECFQ scores. While mediation and moderation analyses have largely been considered to reflect causal relationships, there is support in using these models within cross-sectional, non-causal research with careful interpretations towards the limitations of this form of research [37]. All analyses, unless otherwise specified, were two-tailed and performed in SPSS 26 at an alpha value of 0.05.

Results

All variables were normally distributed with the exception of positively skewed scores of negative affect and, upon visual examination, trait negative emotion. Mean scores and standard deviations are presented in Table 1. Gender and ethnicity differences for all study variables were examined. Independent samples t-tests revealed that females (M=51.386, SD=19.500) tended to score higher than males (M= 46.612, SD=19.142) on global schizotypy, t(303)= −2.130, p=0.034. Similarly, females (M=17.858, SD=8.665) had higher scores of positive schizotypy than males (M=15.051, SD=9.0587), t(303)=−2.717, p=0.007. Furthermore, females (M= 61.189, SD=17.257) scored higher than males (M=55.584, SD=16.616) on levels of cognitive failures, t(303)= −2.858, p=0.005. Mann-Whitney U tests revealed that females (mean rank=168.96) scored higher than males (mean rank=141.62) on negative trait emotion (U=9276.500, p=0.008). Females (mean rank=165.72) also scored higher than males (mean rank=143.93) on negative affect (U=9688.00, p=0.033). One-way ANOVAs revealed a significant difference between ethnicities on cognitive failures F(5, 300)=2.289, p=0.046, and positive schizotypy scores, F(5, 300)=2.504, p=0.031, however, Bonferroni post-hoc analyses revealed no significant differences between specific ethnicities. To examine whether our findings were related to potential response biases (e.g., tendency to report arbitrarily high values across measures), a post-hoc chi-square analysis was performed on the 10^th and 90^th percentiles of the variables of interest. These groups were statistically different from each other (χ²(1, N=306) = 27.00, p<0.001).

Pearson and Spearman correlations between variables of interest are presented in Table 1. Consistent with previous findings [1], global and subscale (i.e., positive, negative, and disorganized) schizotypy scores were positively correlated with scores of cognitive failures, negative trait emotion, and negative affect. Extending previous findings to include positive state emotion, correlations revealed that global and negative schizotypy scores were negatively related to scores of positive affect. Furthermore, negative trait emotion and negative affect scores were positively correlated to cognitive failures, while positive affect scores were negatively correlated to scores of cognitive failures.

Through hierarchical regression analyses, utilizing the Baron and Kenny method [35], we examined the potential moderation effect of negative trait emotion and state emotion (positive and negative affect) on the relationship between global and subscale schizotypy and cognitive failures (shown in Table 2). Upon controlling for gender, negative trait emotion and negative affect moderated the relationship between positive schizotypy and cognitive failures. Decomposition of the interaction revealed that individuals with higher scores of positive schizotypy in addition to higher negative state and traits emotions were more likely to have increased scores of cognitive failures with the following range of t values, t=3.42−11.11, p<0.01. No other moderation effects were observed.

Table 2:

Positive schizotypy and cognitive failures moderated by negative trait and state mood

Positive Schizotypy and Negative Trait Mood			Positive Schizotypy and Negative State Mood
Predictor	R2Δ	β	Predictor	R2Δ	β
Step 1	0.037***		Step 1	0.037***
Gender		0.191***	Gender		0.191***
Step 2	0.335***		Step 2	0.283***
SPQ-BR-P		0.385***	SPQ-BR-P		0.453***
DASS		0.315***	PANAS-N		0.181***
Step 3	0.018**		Step 3	0.009*
SPQ-BR-PxDASS		−0.152**	SPQ-BR-PxPANAS-N		−0.099*

Note. SPQ-BR-P=positive schizotypy, DASS=depression, anxiety, stress scale total score, PANAS-N=negative affect.

^*denotes p<0.05;

^**denotes p<0.01;

^***denotes p≤0.001.

Replicating previous findings [1], negative trait emotion mediated the relationship between schizotypy (global and subscales) and cognitive failures (shown in Fig. 1). Specifically, for the indirect effect (ab=0.087, boot SE=0.001), the bias-corrected bootstrap 95% confidence interval did not cross zero (0.040–0.138), supporting a significant mediating effect of negative trait mood on the relationship between global schizotypy and cognitive failures. Similar indirect effects were present with negative trait mood meditating the relationship between positive (ab= 3.692, boot SE=0.676, 95% CI 2.480–5.104), negative (ab=12.809, boot SE= 0.650, 95% CI 2.749–5.291), and disorganized (ab=2.249, boot SE= 0.523, 95% CI 1.287–3.343) schizotypy and cognitive failures. Total effects for these models with negative trait emotion serving as a mediator were as followed: global schizotypy (c= 0.496, SE=0.046, p=0.077), positive schizotypy (c=14.308, SE=1.316, p<0.001), negative schizotypy (c=10.348, SE=1.140, p<0.001), and disorganized schizotypy (c=13.302, SE=0.905, p<0.001).

Fig. 1.

negative state and trait mood mediations as they relate to cognitive failures and schizotypy dimensions. * denotes p<0.05; ** denotes p<0.01; *** denotes p<0.001; **** denotes p<0.0001.

Additionally, negative affect mediated the relationship between all schizotypy subscales and cognitive failures. Indirect and total effects, respectively, were as followed for positive (ab= 0.129, boot SE=0.038, 95% CI 0.055–0.204; c=1.022, SE=0.094, p<0.001), negative (ab=0.167, boot SE= 0.046, 95% CI 0.083–0.265; c=1.035, SE=0.114, p<0.001), and disorganized (ab= 0.130, boot SE=0.041, 95% CI 0.053–0.214; c= 1.663, SE=0.113, p<0.001) schizotypy. However, negative affect did not mediate the relationship between global schizotypy and cognitive failures; post-hoc analyses confirmed this pattern. Contrary to predictions, positive affect did not mediate the relationship between schizotypy (global and subscales) and scores of cognitive failures.

A post hoc hierarchical linear regression model was conducted to better understand the unique contribution of global schizotypy, as subscales yielded similar findings, above and beyond negative trait emotion in the prediction of cognitive failures. Both schizotypy and negative trait emotion contributed to the prediction of cognitive failures. After controlling for negative trait emotion (β=0.155, p<0.004), global schizotypy (β=0.548, p<0.001) significantly predicted cognitive failures (R²Δ=0.198, FΔ (1, 302) =108.635, p<0.001).

Discussion

Increased cognitive failures may contribute to a more debilitating and distracted life, amplified by negative emotion [38, 39], especially in those with higher levels of schizotypy [1]. We sought to confirm previously established negative trait mood-related findings and expand these prior works to include state emotions such as positive and negative affect.

Confirming previous findings [1], negative trait emotion mediated the relationship between schizotypy and cognitive failures which provides some support for the role of sustained negative emotion in those with elevated levels of schizotypy. Additionally, through our extension of previous findings, we found that state negative emotion similarly mediated the relationship between schizotypy and cognitive failures. While the present study is cross-sectional in nature with limits in causal interpretations, these findings provide some evidence that both enduring and momentary instances of negative emotion impact cognitive failures and subclinical presentations. However, contrary to predictions, the novel inclusion of positive state emotion did not buffer the relationship between cognitive failures and schizotypy. The absence of this relationship is in contrast to previous literature where higher endorsements of positive affect might only be protective for the specific cognitive failure of distractibility [39]. One possible interpretation may be that negative emotions more readily influence both higher levels of schizotypy [27, 23] and cognitive failures [27, 23, 1] than positive emotions. Our findings potentially reflect a negativity bias, where individuals weigh negatively valenced life emotions and emotion more than positively valenced experiences [40, 41], which may be more salient for college-aged students than older adults [42]. Consistent with the stress cascade model for psychosis [43], these cognitive experiences and negative emotions may interact with pre-existing vulnerabilities (e.g., genetic, neurodevelopmental) and stressful college environments, which could theoretically contribute to an increased risk for developing psychosis for some individuals.

The relationship between negative emotion and cognitive failures found in the current study, albeit correlational in nature, might be understood from a ‘resource allocation’ perspective. For instance, negative emotion may be further limiting already taxed cognitive processes [14, 15], thereby leading to increased cognitive failures. When tired or amid stressful experiences, college-aged individuals tend to have difficulty with cognitive processes, such as poor attention and memory, which may contribute to increased cognitive failures [39]. This experience may exemplify our understanding of cognitive load theory as individuals are suggested to have finite cognitive processing abilities [13, 14]. The current sample could therefore be experiencing an amplified cognitive load and devoting increased cognitive resources towards more negative emotions and experiences [41]. At a normative level, fleeting cognitive failures could be considered a minor annoyance. The current findings suggest that the addition of increased negative emotion and schizotypy appear to both independently and reciprocally contribute to increased cognitive failures. As such, there may be value to examining unique and shared contributions to these impairments in everyday living. For example, if an individual routinely misses appointments with supervisors, then these cognitive failures may eventually compound into more severe consequences, possibly fostering additional negative emotions and difficulties. If future research further confirms this relationship, then that might imply that emotion regulation strategies, as well as strategies to limit the emergence of cognitive failures (e.g., structured schedules; digital reminders) may help to reduce impairments related to cognitive failures.

Inconsistent with prior work [1], the current study found a moderation in addition to the mediation. Although this relationship is non-causal, this may suggest that only individuals with higher levels of positive schizotypy and higher levels of negative emotion appeared to have elevated instances of cognitive failures. As such, it appears that individuals predisposed to experiencing negative emotion may be more vulnerable to developing positive symptoms (i.e., hallucinatory experiences), which has been suggested to reflect cognitive resource misallocation. However, negative emotion is likely a non-specific risk contributor towards the emergence of not only schizophrenia-spectrum phenomena, but also other psychopathology (i.e., multifinality) [44]. By assessing both trait-like and state phenomena such as cognitive failures and emotion, we are at best providing estimates, given the influence of state and contextual factors. Thus, cognitive failures themselves can be influenced by both trait functioning patterns and state factors such as mood and time of day [7]. Taken together, these findings suggest negative emotion plays a complex but vital role in the relationship between schizotypy and cognitive failures.

Limitations and strengths

While the use of a convenience sample of undergraduate students may have limited generalizability, we have replicated prior findings Carrigan & Barkus [1] in a similar sample in two different, albeit English-speaking, countries. Levels of schizotypy alone as derived from a self-report measure are not a definitive premorbid indicator but examining dimensional schizotypy in a college age sample may still reveal key insights into the “window of risk.” Similarly, our measured emotion may not fully capture state and trait differences given the use of cross-sectional, single-moment capture of more enduring characteristics. The use of mediations and moderations in the present study are not equated to causality, but rather should be considered within the constraints of the correlational nature of the data. Our post hoc analyses revealed variability in individual scores across the range of measures suggesting that our findings are not solely an artifact of response bias. However, we are sensitive to the potential that these findings need to be interpreted in the context of self-report as the only form of measurement. Lastly, there has been a growing debate pitting subjective and objective measurements of cognitive functioning, stemming from the tendency of individuals with schizotypy reporting increased subjective impairments but with little objective dysfunction. Thus, while our measure of cognitive failures may not be in line with objective functioning, its behavioral considerations may be more salient and readily distressing to the individual. As such, this subjective-objective disjunction [45] should not discount the importance of understanding subjective cognitive failures as they are similar to daily real-world deficits.

Future directions

We demonstrated that trait and, more specifically, state negative emotion appears to contribute to subclinical presentations within the schizophrenia spectrum and the presence of cognitive failures. Therefore, furthering our understanding of the contribution of negative emotion to cognitive failures in schizotypy may provide avenues for eventual interventions. Examining the influences of different conceptualizations of schizotypy such as the inclusion of taxonic, categorical models, in addition to dimensional approaches, will allow future research to more broadly identify and understand the relationship between schizotypy, negative emotion, and cognitive failures. Similarly, including a variety of samples (e.g., community and clinical samples) in addition to subjective and objective forms of measures can allow for a broader consideration of the phenomena of interest. Future research should also continue to parse the unique and shared contributions of schizotypy and negative emotion on the presentation of cognitive failures to better understand nuanced symptom presentations. Additional research is needed to longitudinally examine instances of subjective and objective cognitive failures as they relate to real-life functioning to increase ecological validity. For instance, researchers could capitalize on the use of mobile phone and smartwatch technology and collect temporally precise and ecologically valid measures of cognitive failures, mood states, subclinical phenomena, and symptom presentations. With the expansion of such research, interventions could be developed by focusing on emotion recognition in concert with cognitive treatments to decrease impairments related to cognitive failures in schizophrenia-spectrum conditions.

Conclusions

The present study found evidence to support the role of trait and state emotion on the relationship between schizotypy and cognitive failures. We replicated previous findings involving trait negative emotion and extended the literature by contributing the role of state emotion. To our knowledge, this is among the first studies that include both state and trait emotion as it relates to cognitive failures and schizotypy. As the specific nature of emotion on cognitive failures remains unclear, additional research is warranted to further explore the influence of enduring and momentary emotion. Our findings relating to state emotion provides a unique opportunity for preventive and proactive interventions.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, [TJD], upon reasonable request.