Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jan 1.
Published in final edited form as: Arch Suicide Res. 2022 Nov 14;28(1):123–140. doi: 10.1080/13811118.2022.2137445

Ecological momentary assessment of social approach and avoidance motivations in serious mental illness: Connections to suicidal ideation and symptoms

Emma M Parrish 1, Samantha Chalker 2,3, Mayra Cano 2,4, Philip D Harvey 5, Charles T Taylor 2, Amy Pinkham 6, Raeanne C Moore 2, Robert A Ackerman 6, Colin A Depp 2,3
PMCID: PMC10183051  NIHMSID: NIHMS1865761  PMID: 36377277

Abstract

Objective:

People with serious mental illness (SMI) are at an increased risk of suicide. Social approach and avoidance motivations are linked to social functioning, and social isolation is a risk factor for suicide. This study uses ecological momentary assessment (EMA) to understand social approach and avoidance motivations in relation to symptoms and suicidal ideation (SI).

Method:

Participants (N=128) diagnosed with schizophrenia, schizoaffective disorder, or a mood disorder with psychotic features completed assessments of SI and symptoms at baseline. They completed EMA surveys 3x/day for 10 days. EMA surveys included questions about approach and avoidance motivations, and psychotic symptoms. Participants were split in to four groups based on the median scores of approach and avoidance.

Results:

Participants with SI at baseline had higher mean social avoidance motivation, t(126)=2.84, p=.003, and lower mean social approach motivation, t(126)=−2.44, p=.008, than participants without baseline SI. Greater baseline positive symptoms were related to greater mean avoidance, r=.231, p=.009 but not approach motivation. The low approach/high avoidance group had significantly higher current SI than those with high approach/low avoidance (p<.001). Overall, the low approach/high avoidance group reported more EMA-measured voices than the low approach/low avoidance group (p<.001), and the high approach/low avoidance group (p<.001). Similarly, the low approach/high avoidance group reported more EMA-measured suspiciousness than the low approach/low avoidance (p <.001) and the high approach/low avoidance groups (p <.001).

Discussion:

The results of this study point to the role of social approach and avoidance motivations in relation to SI and psychotic symptoms. Clinically, exposure therapies and cognitive behavioral therapies may help to address these social approach and avoidance processes linked to SI.

Keywords: suicidal ideation, serious mental illness, ecological momentary assessment, social approach motivation, avoidance, psychosis

Introduction

Despite a high rate of suicidal ideation (SI) and behavior in psychotic disorders, the understanding of suicide in psychosis is sparse compared to other illnesses like affective disorders (Aleman & Denys, 2014; Donker et al., 2013; Villa, Ehret, & Depp, 2019). There appear to be some unique phenomenological aspects of suicide linked to psychosis when compared to persons without psychosis, including a higher and faster rate of transition from ideation to behavior, use of different means, and more severe attempts (Chapman et al., 2015; DeVylder, Lukens, Link, & Lieberman, 2015; Kelleher et al., 2013; Lopez-Morinigo et al., 2016). Interpersonal factors (e.g., burdensomeness, belongingness) are prominent in models of suicidal ideation in the general population as well as those pertaining to persons with psychosis (Barzilay & Apter, 2014; Kleiman & Liu, 2013; Klonsky & May, 2015; Parrish et al., 2021; Van Orden et al., 2010; Villa et al., 2019). Disconnection from others and active periods of social withdrawal are also believed to accompany and accelerate the transition from SI to behavior in people with and without psychosis (Calati et al., 2019; Radomsky, Haas, Mann, & Sweeney, 1999). However, the motivational aspects underlying social disconnection as well as the related symptoms and behavioral correlates may be unique. This could have implications for how suicide prevention interventions can be personalized or adapted to people with psychosis.

While social contributors to SI and behavior have received little study in psychosis, diminished social approach motivation is central to understanding the social dysfunction in schizophrenia and related psychoses (Fulford, Campellone, & Gard, 2018; Green et al., 2017). Social approach motivation, which is linked to the positive valence (approach/appetitive) system (Gable & Berkman, 2013) incorporates social goals that focus on approaching and engaging with other people (e.g., meeting up with a friend, introducing oneself to a new person; Gable, 2006; Taylor, Pearlstein, & Stein, 2020). Diminished effort, reward learning, and value representation is typically modelled through tasks and scales, and has been linked with reduced pursuit of rewards (Reinen et al., 2014; Strauss, Waltz, & Gold, 2013), including social rewards (Fulford et al., 2018). Similarly, social anhedonia is associated with reduced social functioning (Barkus & Badcock, 2019; Blanchard, Mueser, & Bellack, 1998; Tan, Shallis, & Barkus, 2020). Given that anhedonia is a negative symptom, it is not surprising that diminished social approach is linked to severity of negative symptoms (Reddy et al., 2014; Reddy, Horan, & Green, 2016). Yet, with respect to suicide, negative symptoms appear to have a complicated relationship. In two studies of social anhedonia in psychotic disorders, no association or a negative association with SI or suicidal behavior was found (Jahn et al., 2016; Loas, Azi, Noisette, Legrand, & Yon, 2009). This stands in stark contrast to the role of social anhedonia as a strong risk factor for suicide in depression, over and above other depressive symptoms (Bonanni et al., 2019). However, one large study of negative symptoms and suicide among Veterans with schizophrenia found relationships between individual negative symptom domains and suicide (Harvey et al., 2018). Namely, blunted affect and alogia were associated with an increased risk of suicidal behavior, whereas avolition was associated with a decreased risk of suicidal behavior (Harvey et al., 2018). Thus, the relationship of negative symptoms, particularly social anhedonia, to SI and behavior requires further clarification.

Compared to diminished social approach motivation, social avoidance motivation is less studied in psychosis as a predictor of social dysfunction (Fulford et al., 2018). Social avoidance, which has been linked to the negative valence (avoidance/aversive) system (Gable & Berkman, 2013), is associated with goals to avoid others (e.g., the desire to actively stay away from others for fear of negative social outcomes; Gable, 2006; Taylor et al., 2020). Social avoidance has been linked to positive symptoms, particularly persecutory delusions, among people with schizophrenia (Freeman, Taylor, Molodynski, & Waite, 2019). Yet, a recent review reveals a compelling body of literature supporting that components of social avoidance (e.g., social threat, anxiety) may have a significant role in social dysfunction in psychosis (Fulford et al., 2018). In our research group, we found leaving the home is associated with increases in anxiety in people with schizophrenia (Parrish et al., 2020). To our knowledge, however, few or no studies have evaluated avoidance as a predictor of SI or behavior in psychosis. Furthermore, social approach and avoidance motivations could synergistically impact social dysfunction. Social approach and avoidance motivations are partially independent dimensions that have been shown to impact social functioning through different processes (Elliot, Gable, & Mapes, 2006; Gable, 2006). It is therefore possible for people to vary in different ways along each dimension (e.g., both high approach and avoidance motivations; high avoidance and low approach motivations), which may relate to psychotic symptoms and contribute to suicidal ideation.

Ecological momentary assessment (EMA) provides a potential way to evaluate the shared and unique concomitants of approach and avoidance motivations, since both opposing motivations can be modelled in tandem over time within individuals. A number of EMA studies in schizophrenia have evaluated social approach and avoidance motivations in relation to affect and behavior (Badal, Parrish, Holden, Depp, & Granholm, 2021; Granholm, Ben-Zeev, Fulford, & Swendsen, 2013; Parrish et al., 2020), but the relationship of these constructs to suicide is not well understood. Indeed, a recent systematic review examining EMA studies of suicidal ideation and behavior in broad clinical populations commented on the lack of studies that use EMA to understand social interactions in their relationship to suicide (Gee, Han, Benassi, & Batterham, 2020). To our knowledge, no EMA studies have simultaneously included both social approach and avoidance motivations as separate constructs, nor examined their relation to SI.

In this paper, we evaluate EMA queries of future near term social approach motivation (i.e., How much interest or motivation do you have in interacting with others later today?) and social avoidance motivation (i.e., How much do you want to avoid others later today?) in a sample of people with schizophrenia, schizoaffective disorder, or a mood disorder with psychotic features. We focused on the relationship between approach and avoidance and 1) presence/severity of current SI and past suicide attempts 2) social behavior (i.e., surveys reported as being alone), and 3) psychotic symptoms, depressive symptoms, and affect. Aims 1 and 2 examined aggregated EMA data, whereas aim 3 was examined with both aggregated and momentary EMA data. We predicted that diminished social approach motivation and increased social avoidance motivation would be independently associated with current SI and past suicide attempts, more time spent alone, and more severe psychotic and depressive symptoms. Given the link between diminished social approach motivation and anhedonia (Germans & Kring, 2000), we hypothesized that diminished positive affect and greater clinically rated depression would be more strongly associated with diminished social approach than avoidance motivation. Furthermore, considering the strong link between positive symptoms and avoidance (Freeman et al., 2019), we hypothesized that psychotic symptoms would be more strongly associated with social avoidance than approach motivation. Finally, in an exploratory analysis we examined patterns of convergence of diminished social approach and social avoidance motivations on current SI, psychotic symptoms, and depressive symptoms.

Method

Participants.

Data were from the ongoing Social Cognition and Self-Harm in Mental Illness Study, and the full study protocol has been previously detailed elsewhere (Parrish et al., 2021). For the current study, we included data from 128 adults aged 18 and older with a diagnosis of schizophrenia (n=43), schizoaffective disorder (n=56), or affective disorder with psychotic features (n=29). Inclusion criteria included: 1) aged 18–65; 2) current diagnosis of schizophrenia, schizoaffective disorder, bipolar disorder with psychotic features, or major depressive disorder with psychotic features, confirmed by the Mini International Neuropsychiatric Inventory (MINI; Sheehan et al., 1998) and the Structured Clinical Interview for the DSM-5 psychotic symptoms subsection (SCID-5; First et al., 2015); 3) available informant with whom the participant was regularly in contact; 4) in outpatient, partial hospitalization, or residential care; 5) proficient in English; and 6) able to provide informed consent. Participants were excluded if they had: 1) history of a head trauma with loss of consciousness >15 min; 2) diagnosed with neurological or neurodegenerative disorder; 3) vision or hearing problems that would interfere with data collection; 4) estimated IQ < 70, as determined by the Wide Range Achievement Test-4 (WRAT-4; Wilkinson and Robertson, 2006); 5) DSM-5 diagnosis of a substance use disorder in the past three months, excluding cannabis and tobacco, confirmed by the SCID-5.

Participants completed a lab-based baseline visit followed by a 10-day smartphone EMA protocol. Before the COVID-19 pandemic, participants’ eligibility and symptoms were assessed in-person either in the lab space or in the community (n=102). Due to social distancing protocols put in place during the COVID-19 pandemic, 26 participants were assessed remotely via telephone. Because of the remote nature of these assessments, research assistants could not collect data on a Positive and Negative Syndrome Scale (PANSS) item that required visual observation (N1, Blunted Affect).

Measures.

Depression.

The 10-item Montgomery-Åsberg Depression Rating Scale (MADRS; Montgomery and Åsberg, 1979) was used to assess the severity of depression. The interview-administered scale consists of 10 items, each of which is scored from 0 (symptom not present or normal) to 6 (severe or continuous presence of the symptom), for a total possible score of 60. Higher total scores indicate higher severity of depressive symptoms.

Positive and negative symptoms.

Severity of psychotic symptoms was evaluated with the Positive and Negative Syndrome Scale (PANSS; Kay et al., 1987). Trained raters administered the semi-structured interview in its entirety on the same day as the MADRS. The PANSS consists of 30 items scored on a seven-point Likert scale ranging from 1 to 7. The items are further differentiated on three subscales: seven items comprise the positive symptoms scale (P1–P7) and seven items comprise the negative symptoms scale (N1–N7). The range of the sub-scales is 7 to 49. In addition to the positive and negative symptoms subscales, we also created a subscale to represent a subdomain of negative symptoms, emotional experience (emotional withdrawal – N2, passive social withdrawal – N4, active social avoidance – G16; Harvey, Khan, & Keefe, 2017).

SI and suicide attempts.

SI was collected and analyzed in two ways: categorically based on the presence or absence of SI, and continuously based on SI severity in the past 48 hours.

First, presence of current (past month) and past (lifetime) SI and its severity was assessed by the semi-structured Columbia-Suicide Severity Rating Scale (C-SSRS; Posner et al., 2011). The C-SSRS was also used to assess recent (past 3 months) and lifetime suicidal behavior, including suicide attempts. The SI scale defines five levels of suicidal ideation: 1-Wish to be dead; 2-Non-Specific Active Suicidal Thoughts; 3-Active SI with Any Methods (Not Plan) without Intent to Act; 4-Active SI with Some Intent to Act, without Specific Plan; 5-Active SI with Specific Plan and Intent. Study participants were grouped into either no current SI (SI−) or current SI (SI+), as defined by a rated score of two (i.e., non-specific active suicidal thoughts) or higher within the last month. Participants self-reporting a recent suicide attempt (i.e., “a potentially self-injurious act committed with at least some wish to die as a result of act”; Posner et al., 2011) within the last 3 months were also grouped into the current SI (SI+) group. In this sample, all participants who reported a recent suicide attempt also reported SI in the past month.

Second, the Modified Scale for Suicidal Ideation (MSSI, Millet et al., 1986) is an 18-item semi-structured interview used to assess SI severity in the past 48 hours and was used in the current study as a dimensional measure of suicidal ideation severity. Each item is rated from 0 to 3 and the total score ranges from 0 to 54, with higher scores representing greater suicide ideation severity. Domains include intensity of attitudes, competence to attempt, and acts of preparation to attempt.

EMA Procedure.

Participants were given the choice to use their own smartphones or use a lab-provided Samsung A50 smartphone for the EMA surveys. EMA surveys were administered using the NeuroUX platform (https://www.getneuroux.com/). During the baseline visit, either in-person or remotely, participants were trained in procedures and were asked to complete an in-lab EMA survey to confirm knowledge of surveys. Participants chose three daily sampling windows in the morning, afternoon, and evening that were at least 2 hours apart, during which they would be available to receive text messages with weblinks to EMA surveys 3 times a day across the 10-day period.

Surveys were randomly delivered within the previously chosen sampling windows 3x/day for 10 days, for a total of 30 possible EMA samples. Participants were instructed to complete EMA surveys within 1 hour of receiving them. All responses were time-stamped. On a one to seven Likert-type scale, participants rated the current presence and intensity of experiences of voices, mistrust, and mood (happy, sad). They rated their level of social approach and avoidance motivations for later in the day (Social approach: How much interest or motivation do you have in interacting with others later today? Social avoidance: How much do you want to avoid others later today?) on a one to seven Likert-type scale. For all EMA measures, higher ratings represent higher levels of the measured construct. Participants also noted who they were with and who they had previously interacted with, categorized as “alone” or “with others.” Participants were paid $1.66 for each survey completed, for a maximum of $50. To encourage adherence, participants were also contacted by telephone on the first day of EMA to troubleshoot and contacted again if they missed more than three consecutive surveys.

Statistical analyses.

All analyses were performed using SPSS v.28. Descriptive statistics were calculated for all demographic characteristics. Because an item on the PANSS, N1, was missing for 26 participants due to remote data collection, this variable was imputed using multiple imputation monotone method (100 imputations) and entering all other N symptoms (N2-N7) to inform this analysis. The average N1 score for these 100 imputations was substituted into the PANSS negative symptom total. All analyses involving the PANSS negative symptom total were performed with and without the imputed N1 item, and statistical significance of the findings did not change with the removal of this imputed variable. Thus, all analyses reported here that include the PANSS negative symptom total include the imputed N1 item. An independent t-test was performed to compare mean levels of social approach and avoidance motivations between participants with and without current SI, and a Pearson correlation was performed between SI severity, aggregated social avoidance motivation, and aggregated social approach motivation across the EMA period. We ran a centered mixed effects model (Twisk, 2019) to understand how EMA-measured social approach and avoidance motivations related to EMA psychotic symptoms. Predictors for all models included a mean EMA value of the item (future social approach or avoidance motivations) over time for each participant, each participant’s momentary deviation from the mean at each EMA survey, and a random intercept for each participant. For example, using this centering technique, if the avoidance momentary variable is positive, that means that the participant reported experiencing more avoidance at that survey than their mean across the full EMA period. Dependent variables included EMA-reported voices, mistrust, and mood (happy, sad). Participants were split in to four groups based on the median of social approach and avoidance motivations: low approach/high avoidance; low approach/low avoidance; high approach/low avoidance; and high approach/high avoidance. Rater-assessed SI in the past 48 hours (MSSI) based on these 4 groups in aggregate was compared using a one-way ANOVA and a post hoc Tukey test. Finally, four mixed effects linear models were performed to understand the impact of these four groups on EMA-measured voices, mistrust, and mood (happy, sad), using a pairwise comparison to compare the low approach/high avoidance group to the other three groups. Due to multiple comparisons and a high probability of type 1 error, P was set to .01 for all analyses, which is more conservative than utilizing a false discovery rate.

Results

Demographic characteristics.

See Table 1 for full statistics. Participants had an overall mean age of 43.4, over half (56.3%) were female, 45.3% identified as Black or African American, and had an average of 12.8 years (SD=2.3) of education. Participants with and without SI were well matched on all demographic characteristics, with no significant differences in age, sex, race, ethnicity, education, or employment (p’s>.01). Baseline data were as expected, such that participants with SI had higher scores of MADRS depression than participants without SI (t(125)=6.0, p<.001), had more lifetime suicide attempts (t(126)=2.4, p=.009), and had higher SI severity in the past 48 hours (t(126)=4.44, p<.001). There were no differences between the two groups on any other symptom measure, nor EMA survey adherence (p’s>.01).

Table 1.

Sample Characteristics

Characteristic SI+ (N=69) SI− (N=59) Total Sample (N=128) t or X2, p
Age, M(SD), Range 43.5 (12.0), 19–64 43.4 (10.8), 22–65 43.4 (11.4), 19–65 .01, .496
Sex (female), n (%) 38 (55.1%) 34 (57.6%) 72 (56.3%) .084, .771
Race, n (%)
 White or Caucasian 20 (29.0%) 20 (33.9%) 40 (31.3%) 8.42, .015
 Black or African American 26 (37.7%) 32 (54.2%) 58 (45.3%)
 Other 23 (33.3%) 7 (11.9%) 30 (23.4%)
Ethnicity (Hispanic), n (%) 17 (24.6%) 10 (16.9%) 27 (21.1%) 1.13, .288
Education (years), M(SD), Range 12.9 (2.6), 4–18 12.8 (1.9), 9–16 12.8 (2.3), 4–18 .06, .476
Employment Status, n (%)
 Employed or In School (full or part time) 21 (30.4%) 13 (22.0%) 34 (26.6%) 1.15, .283
 Not Employed 48 (69.6%) 46 (78.0%) 94 (73.4%)
 Diagnosis, n (%)
 Schizophrenia 17 (24.6%) 26 (44.1%) 43 (43.8%)
 Schizoaffective 34 (49.3%) 22 (37.3%) 56 (43.8%) 5.52, .137
 Bipolar disorder with psychotic features 17 (24.6%) 10 (16.9%) 27 (21.1%)
 Major depressive disorder with psychotic features 1 (1.4%) 1 (1.7%) 2 (1.6%)
MCCB Age-Corrected T Scores a, M(SD), Range
 Processing Speed 42.7 (13.3), 13–68 42.1 (11.1), 12–65 42.5 (12.2), 12–68 .26, .397
 Working Memory 38.9 (10.4), 13–62 41.5 (7.5), 27–56 40.1 (9.2), 13–62 −1.57, .059
 Verbal Learning 38.8 (9.3), 24–67 39.7 (9.0), 26–72 39.2 (9.1), 24–72 −.55, .293
PANSS Positive Symptoms, M(SD), Range 18.1 (5.6), 7–34 17.5 (5.6), 7–31 17.8 (5.6), 7–34 .61, .273
PANSS Negative Symptoms, M(SD), Range 13.48 (4.20), 7–26 12.77 (3.65), 7–24 13.1 (4.0), 7–26 1.01, .156
PANSS Emotional Experience Subscale, M(SD), Range 6.7 (2.6), 3–14 6.4 (2.7), 3–13 6.6 (2.7), 3–14 0.48, .318
MADRS, M(SD), Range 20.8 (11.4), 0–39 9.6 (9.2), 0–31 15.6 (11.8), 0–39 6.0, <.001*
YMRS, M(SD), Range 5.0 (6.1), 0–19 3.0 (5.8), 0–23 4.1 (6.0), 0–23 1.9, .029
C-SSRS-Lifetime Suicide Attempt, n (%) 3.9 (5.0), 0–20 1.9 (4.5), 0–30 3.0 (4.9), 0–30 2.4, .009*
MSSI – Suicidal Ideation Severity in the Past 48 Hours, M(SD), Range 4.7 (8.0), 0–30 0.1 (0.4), 0–2 2.6 (6.3), 0–30 4.44, <.001*
EMA Survey Adherence Percentage, M(SD), Range 80.1 (19.4), 10–100 79.7 (22.6), 10–100 79.9 (20.9), 10–100 .129, .449
Open in a new tab

Note. MCCB = MATRICS Consensus Cognitive Battery; PANSS = positive and negative syndrome scale; MADRS = Montgomery-Asberg Depression Rating Scale; YMRS = Young Mania Rating Scale; C-SSRS = Columbia-Suicide Severity Rating Scale; MSSI = Modified Scale for Suicidal Ideation.

*

indicates significant at p<.01

Descriptive statistics of EMA variables.

See Table 2 for full statistics. When EMA scores were aggregated, participants with current SI had lower social approach motivation (t(126)=−2.44, p=.008), higher social avoidance motivation (t(126)=2.84, p=.003), lower happiness (t(126)=−3.36, p<.001), and higher sadness (t(126)=2.52, p=.007) than participants without current SI. There were no differences in mean EMA ratings between participants with and without SI on EMA-measured voices (t(126)=1.22, p=.112) or suspiciousness (t(126)=1.70, p=.046). In terms of diagnostic differences, participants with schizophrenia-spectrum disorders reported greater mean levels of voices than participants with mood disorders with psychotic features (t(73.66)=4.76, p<.001). There were no other mean diagnostic differences on EMA variables (p’s>.082). In the full sample, mean social approach motivation and mean social avoidance motivation were highly negatively correlated (r=−.736, p<.001). Additionally, there was no correlation between number of lifetime suicide attempts and mean social approach motivation (r=−.107, p=.230) or mean social avoidance motivation (r=.135, p=.129).

Table 2.

EMA Average Responses

Variable Question (Since the past alarm…) SI+; M(SD), [Range] (N=69) SI−; M(SD), [Range] (N=59) T, p Cohen’s D
Social Approach Motivation How much interest or motivation do you have in interacting with others later today? 3.82 (1.46), [1.00, 6.93] 4.44 (1.40), [1.0, 7.0] −2.44, .008* 0.43
Social Avoidance Motivation How much do you want to avoid others later today? 4.12 (1.46), [1.20, 6.79] 3.33 (1.66), [1.00, 7.00] 2.84, .003* 0.49
Voices How much have you been bothered by voices? 2.50 (1.61), [1.00, 6.92] 2.16 (1.45), [1.00, 6.13) 1.22, .112 0.22
Suspiciousness How much have you had thoughts that you really can’t trust other people? 3.65 (1.60), [1.00, 7.00] 3.15 (1.72), [1.00, 7.00] 1.70, .046 0.30
Happy How much have you felt happy? 3.87 (1.36), [1.00, 7.00] 4.71 (1.45), [1.30, 7.00] −3.36, <.001* 0.60
Sad How much have you felt sad or depressed? 3.31 (1.47), [1.00, 6.79] 2.66 (1.46), [1.00, 6.65] 2.52, .007* 0.44
Open in a new tab

Note.

*

p<.01

Do social approach and avoidance motivations (aggregate EMA) relate to suicide ideation?

Higher MSSI scores, which reflect higher severity of SI in the past 48 hours, negatively correlated with higher mean social approach motivation (r=−.289, p<.001) and were positively correlated with higher mean social avoidance motivation (r=.282, p=.001). Mean social approach (t(126)=−1.13, p=.129) and mean social avoidance (t(126)=.27, p=.396) motivations did not differ between participants who did and did not have a history of suicide attempt, and were not significantly correlated with number of lifetime suicide attempts (p’s>.129).

Do social approach and avoidance motivations (aggregate EMA) relate to baseline symptom severity and social behavior?

Higher PANSS positive symptoms were related to higher social avoidance motivation (r=.231, p=.009). However, PANSS positive symptom severity was not related to social approach motivation, and PANSS negative symptom severity was unrelated to both social approach and avoidance motivations. Higher scores on the PANSS emotional experience subscale correlated with lower social approach motivation (r=−.266, p=.002) and higher social avoidance motivation (r=.290, p<.001). Higher depressive symptomatology was correlated with lower social approach motivation (r=−.419, p<.001) and higher social avoidance motivation (r=.380, p<.001). Additionally, social avoidance motivation was associated with less time spent with others (r=−.243, p=.006), but higher social approach motivation was not significantly associated with higher time spent with others (r=.196, p=.026).

Do participants differ on lab-measured SI, depression, and psychotic symptoms based on the combination of social approach and avoidance motivations?

There were 39 participants with low approach/high avoidance, 19 participants with low approach/low avoidance, 45 participants with high approach/low avoidance, and 25 participants with high approach/high avoidance. The four groups differed in SI severity (F(3, 124)=6.29, p<.001) and the low approach/high avoidance group had a significantly higher MSSI score than those with high approach/low avoidance (p<.001, see Figure 1a). Similarly, the four groups differed in their MADRS score (F(3, 123)=11.70, p<.001) and the low approach/high avoidance group had a significantly higher MADRS score than those with high approach/low avoidance (p<.001, see Figure 1b), as well as those with high approach/high avoidance (p<.003, see Figure 1b). The four groups did not differ on PANSS positive (F(3, 124)=1.71, p=.168) or negative symptoms (F(3, 124)=0.88, p=.453), but they did differ on the PANSS emotional experience subscale (F(3, 124)=4.49, p=.005), such that the low approach/high avoidance group had higher PANSS emotional experience scores than the high approach/low avoidance group (p=.003).

Figure 1a:

Figure 1a:

Open in a new tab

One-way ANOVA comparing MSSI total score

Note. All p-values correspond to a post-hoc Tukey test comparison with the low motivation, high avoidance group. *p<.01.

Figure 1b:

Figure 1b:

Open in a new tab

One-way ANOVA comparing MADRS total score

Note. All p-values correspond to a post-hoc Tukey test comparison with the low motivation, high avoidance group. *p<.01.

Do participants’ EMA responses differ based on the combination of social approach and avoidance motivations?

First, we estimated four linear mixed models with voices, mistrust, happiness, sadness as the response variable in each model. Each model included a random intercept for participant, a mean EMA value of social approach and avoidance motivations for each participant, and a momentary deviation from the mean for each participant. See Table 3 for full statistics. Mean levels of social approach and avoidance motivations, along with momentary deviations in social approach and avoidance motivations, were significantly related to happiness (p’s <.009) and sadness (p’s<.001). However, a pattern emerged with voices and mistrust, such that mean levels of social approach and avoidance motivation were significantly related to both voices (p’s<.001) and mistrust (p’s<.002), whereas only momentary deviations in social avoidance motivation was significantly related to voices (p<.001) and mistrust (p<.001). Momentary deviations in social approach motivation were not significantly related to voices (p=.498) or mistrust (p=.100).

Table 3.

Centered Mixed Models

Response Variable Estimate S.E. T P
Happiness Avoidance Momentary −0.14 0.02 −7.10 <.001*
Avoidance Mean 0.06 0.02 2.60 0.009*
Approach Momentary 0.20 0.02 9.78 <.001*
Approach Mean 0.85 0.03 33.21 <.001*
Sadness Avoidance Momentary 0.17 0.02 7.44 <.001*
Avoidance Mean 0.48 0.03 16.93 <.001*
Approach Momentary −0.12 0.02 −4.95 <.001*
Approach Mean −0.11 0.03 −3.68 <.001*
Voices Avoidance Momentary 0.10 0.02 4.20 <.001*
Avoidance Mean 0.63 0.03 21.86 <.001*
Approach Momentary −0.02 0.02 −0.68 0.498
Approach Mean 0.30 0.03 9.84 <.001*
Suspiciousness Avoidance Momentary 0.18 0.03 7.25 <.001*
Avoidance Mean 0.49 0.03 15.64 <.001*
Approach Momentary −0.04 0.03 −1.64 0.100
Approach Mean −0.11 0.03 −3.12 0.002*
Open in a new tab

Note.

*

p<.01

Next the low approach/high avoidance group was compared to the other three groups on EMA variables: voices, mistrust, happiness, and sadness (Figure 2). Overall, the low approach/high avoidance group reported higher severity of voices than the low approach/low avoidance group (p<.001), as well as the high approach/low avoidance group (p<.001). However, the low approach/high avoidance group did not report significantly more voices than the high approach/high avoidance group (p=.374). Similarly, the low approach/high avoidance group reported more mistrust than the low approach/low avoidance group (p<.001) and the high approach/low avoidance group (p<.001), but not the high approach/high avoidance group (p=.018). As for mood, the low approach/high avoidance group reported a lower level of happiness than the three other groups (p<.001), as well as a higher level of sadness (p<.001).

Figure 2:

Figure 2:

Open in a new tab

Linear mixed models comparing EMA variables by motivation and avoidance group

Note. All p-values correspond to a post-hoc Tukey test comparison with the low motivation, high avoidance group. *p<.01.

Discussion

In this study, we found that social approach and avoidance motivations related to SI and behavioral withdrawal in people with psychotic disorders and mood disorders with psychotic features. Social approach and avoidance motivations were highly correlated in this sample, yet had a distinct pattern of correlates, converging with prior literature examining social approach and avoidance motivations in non-clinical samples (Elliot et al., 2006; Gable, 2006). While social approach motivation is more aligned with a happy mood, social avoidance motivation is aligned with a sad mood. Crucially, the combination of social approach and avoidance motivations seem to be especially important in terms of their relationship to SI in the past 48 hours, lab-measured depressive symptoms, and real-time sadness, as unique patterns of these constructs relate differentially to SI and symptoms. While social approach and avoidance motivations were related independently to PANSS positive psychotic symptoms, the combination of these two constructs did not contribute at the aggregate level to differences in PANSS positive symptoms. However, the combination of these constructs related to differences in momentary experiences of voices and mistrust, and it seems that the presence of voices and mistrust increases the relationship of social avoidance motivation and SI.

As research has focused primarily on reward in the study of social approach motivation in schizophrenia (Najas-Garcia, Carmona, & Gómez-Benito, 2018), previous literature notes the need to have additional focus on avoidance processes in schizophrenia (Fulford et al., 2018). The results of this study point to unique patterns of social approach and avoidance motivations, namely convergence of social approach and avoidance motivation constructs, and how they relate to SI and symptoms. The results of this study also highlight the importance of social avoidance processes relating to SI within this population, because of the pronounced pattern of high social avoidance linked to SI. Additionally, the results of this study suggest that social avoidance motivation may relate specifically to psychotic symptoms. The EMA data suggests that social avoidance motivation is especially related to voices and mistrust, due to the lack of significant difference in these variables between the groups that reported high levels of social avoidance motivation. Based on the results of this study, we may speculate that voices and paranoia contribute to SI in their relation to social avoidance motivation, whereas anhedonia and negative mood are linked with social approach motivation. Furthermore, the implication of social avoidance in processes related to SI may be related to greater threat perception, which has been linked to SI in schizophrenia (Depp, Villa, Schembari, Harvey, & Pinkham, 2018; Villa et al., 2018).

These results show that the combination of high social avoidance motivation and low social approach motivation is related to SI in this population. Indeed, previous literature shows that depression is a risk factor for suicide in psychosis (Cassidy, Yang, Kapczinski, & Passos, 2018). However, while social anhedonia relates to suicide among people with depression, this is not the case for people with schizophrenia (Bonanni et al., 2019; Jahn et al., 2016; Loas et al., 2009). Thus, symptoms of diminished social approach and avoidance motivations may particularly relate to suicide in this population. These results may explain why social anhedonia is not an independent risk factor for suicide in psychosis, as perhaps it is in the context of social avoidance motivation.

Clinically, these results point to the importance of both social approach and avoidance motivation processes in SMI, and the connection of these processes to SI. Behavioral activation and other exposure therapies may help people with schizophrenia to become comfortable around others, confronting social avoidance motivation and potentially increasing social approach motivation. Additionally, cognitive behavioral therapy for psychosis (CBTp) may help clients gain a greater understanding of their social avoidance processes and challenge thinking surrounding such processes.

Limitations of this study include that these analyses are correlational and do not imply causation (e.g., social avoidance may increase hearing voices, but this conclusion cannot be drawn from this data), and that we did not examine future suicidal behavior. The method of measurement of social approach and avoidance motivations limit a more nuanced analysis on the impact on actual social behavior. The analytic method of exploring profiles of social approach and avoidance motivations was chosen to understand patterns of these related constructs, and future work with more robust methods (e.g., multiple items) would be needed to fully understand the confluence of these constructs. Additionally, other more targeted measures of negative symptoms (e.g., Clinical Assessment Interview for Negative Symptoms, CAINS, Kring et al., 2013; Scale for the Assessment of Negative Symptoms, SANS, Andreasen, 1989) may provide a more sensitive and nuanced picture of negative symptoms than the PANSS. Furthermore, while financial compensation of the participants is common in EMA studies (Sedano-Capdevila, Porras-Segovia, Bello, Baca-García, & Barrigon, 2021) and allowed for high adherence to EMA surveys, it is possible that this financial incentive led to a selection bias, as adherence to EMA surveys is significantly lower when no financial incentive is provided (Lopez-Morinigo et al., 2021). It is important to consider that social avoidance motivation processes may also impact help-seeking among people with schizophrenia (e.g., decreased contact to crisis lines, decreased engagement in treatment), but future studies should directly investigate this. Future research should also explore social withdrawal and isolation as a potential mechanism for the relationship of social approach and avoidance motivations to SI. Mobile interventions, while an exciting avenue for future research, could be explored in relation to social approach and avoidance motivations, though it is important to note that the addition of a digital device may not always be beneficial (Porras-Segovia et al., 2021). Furthermore, future research should investigate the combination of psychosis and mood symptoms as they relate to suicide, especially related to social affiliation.

Acknowledgements:

We would like to thank Katelyn Barone, Bianca Tercero, Cassi Springfield, Linlin Fan, Ian Kilpatrick, Snigdha Kamarsu, Tess Filip, Avery Quynh, Vanessa Scott, and Maxine Hernandez for their involvement in data collection and recruitment.

Funding Information:

This work was supported by the National Institute of Mental Health (grant number: NIMH R01 MH116902-01A1; and T32 MH019934 to E.M.P.).

Footnotes

Declaration of Interest Statement:

R.C.M. is a co-founder of KeyWise AI, Inc. and a consultant for NeuroUX. The terms of this arrangement have been reviewed and approved by UC San Diego in accordance with its conflict of interest policies.

P.D.H. has received consulting fees or travel reimbursements from Alkermes, Bio Excel, Boehringer Ingelheim, Karuna Pharma, Merck Pharma, Minerva Pharma, SK Pharma, and Sunovion (DSP) Pharma in the past year. He receives royalties from the Brief Assessment of Cognition in Schizophrenia (Owned by WCG Verasci, Inc. and contained in the MCCB). He is chief scientific officer of i-Function, Inc.

All other authors have no interests to disclose.

References

  1. Aleman A, & Denys D (2014). Mental health: A road map for suicide research and prevention. Nature, 509(7501), 421–423. [DOI] [PubMed] [Google Scholar]
  2. Andreasen NC (1989). The Scale for the Assessment of Negative Symptoms (SANS): Conceptual and theoretical foundations. Br J Psychiatry Suppl, 155(7), 49–58. (2695141). [PubMed] [Google Scholar]
  3. Badal VD, Parrish EM, Holden JL, Depp CA, & Granholm E (2021). Dynamic contextual influences on social motivation and behavior in schizophrenia: A case-control network analysis. Npj Schizophrenia, 7(1), 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Barkus E, & Badcock JC (2019). A Transdiagnostic Perspective on Social Anhedonia. Frontiers in Psychiatry, 10, 216. 10.3389/fpsyt.2019.00216 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Barzilay S, & Apter A (2014). Psychological models of suicide. Archives of Suicide Research, 18(4), 295–312. [DOI] [PubMed] [Google Scholar]
  6. Blanchard JJ, Mueser KT, & Bellack AS (1998). Anhedonia, Positive and Negative Affect, and Social Functioning in Schizophrenia. Schizophrenia Bulletin, 24(3), 413–424. 10.1093/oxfordjournals.schbul.a033336 [DOI] [PubMed] [Google Scholar]
  7. Bonanni L, Gualtieri F, Lester D, Falcone G, Nardella A, Fiorillo A, & Pompili M (2019). Can Anhedonia Be Considered a Suicide Risk Factor? A Review of the Literature. Medicina, 55(8), 458. 10.3390/medicina55080458 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Calati R, Ferrari C, Brittner M, Oasi O, Olié E, Carvalho AF, & Courtet P (2019). Suicidal thoughts and behaviors and social isolation: A narrative review of the literature. Journal of Affective Disorders, 245, 653–667. 10.1016/j.jad.2018.11.022 [DOI] [PubMed] [Google Scholar]
  9. Cassidy RM, Yang F, Kapczinski F, & Passos IC (2018). Risk Factors for Suicidality in Patients With Schizophrenia: A Systematic Review, Meta-analysis, and Meta-regression of 96 Studies. Schizophrenia Bulletin, 44(4), 787–797. 10.1093/schbul/sbx131 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chapman CL, Mullin K, Ryan CJ, Kuffel A, Nielssen O, & Large MM (2015). Meta-analysis of the association between suicidal ideation and later suicide among patients with either a schizophrenia spectrum psychosis or a mood disorder. Acta Psychiatrica Scandinavica, 131(3), 162–173. 10.1111/acps.12359 [DOI] [PubMed] [Google Scholar]
  11. Depp CA, Villa J, Schembari BC, Harvey PD, & Pinkham A (2018). Social cognition and short-term prediction of suicidal ideation in schizophrenia. Psychiatry Research, 270, 13–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. DeVylder JE, Lukens EP, Link BG, & Lieberman JA (2015). Suicidal ideation and suicide attempts among adults with psychotic experiences: Data from the collaborative psychiatric epidemiology surveys. JAMA Psychiatry, 72(3), 219–225. 10.1001/jamapsychiatry.2014.2663 [DOI] [PubMed] [Google Scholar]
  13. Donker T, Calear A, Grant JB, Van Spijker B, Fenton K, Hehir KK, … Christensen H (2013). Suicide prevention in schizophrenia spectrum disorders and psychosis: A systematic review. BMC Psychology, 1(1), 6. 10.1186/2050-7283-1-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Elliot AJ, Gable SL, & Mapes RR (2006). Approach and avoidance motivation in the social domain. Personality and Social Psychology Bulletin, 32(3), 378–391. [DOI] [PubMed] [Google Scholar]
  15. Freeman D, Taylor KM, Molodynski A, & Waite F (2019). Treatable clinical intervention targets for patients with schizophrenia. Schizophrenia Research, 211, 44–50. [DOI] [PubMed] [Google Scholar]
  16. Fulford D, Campellone T, & Gard DE (2018). Social motivation in schizophrenia: How research on basic reward processes informs and limits our understanding. Clinical Psychology Review, 63, 12–24. (WOS:000439086800002). 10.1016/j.cpr.2018.05.007 [DOI] [PubMed] [Google Scholar]
  17. Gable SL (2006). Approach and avoidance social motives and goals. Journal of Personality, 74(1), 175–222. [DOI] [PubMed] [Google Scholar]
  18. Gable SL, & Berkman ET (2013). 12 making connections and avoiding loneliness: Approach and avoidance social motives and goals. Handbook of Approach and Avoidance Motivation, 203. [Google Scholar]
  19. Gee BL, Han J, Benassi H, & Batterham PJ (2020). Suicidal thoughts, suicidal behaviours and self-harm in daily life: A systematic review of ecological momentary assessment studies. Digital Health, 6, 2055207620963958. 10.1177/2055207620963958 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Germans MK, & Kring AM (2000). Hedonic deficit in anhedonia: Support for the role of approach motivation. Personality and Individual Differences, 28(4), 659–672. [Google Scholar]
  21. Granholm E, Ben-Zeev D, Fulford D, & Swendsen J (2013). Ecological Momentary Assessment of social functioning in schizophrenia: Impact of performance appraisals and affect on social interactions. Schizophrenia Research, 145(1–3), 120–124. 10.1016/j.schres.2013.01.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Green MF, Horan WP, Lee J, McCleery A, Reddy LF, & Wynn JK (2017). Social disconnection in schizophrenia and the general community. Schizophrenia Bulletin, 44(2), 242–249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Harvey PD, Khan A, & Keefe RS (2017). Using the positive and negative syndrome scale (PANSS) to define different domains of negative symptoms: Prediction of everyday functioning by impairments in emotional expression and emotional experience. Innovations in Clinical Neuroscience, 14(11–12), 18. [PMC free article] [PubMed] [Google Scholar]
  24. Harvey PD, Posner K, Rajeevan N, Yershova KV, Aslan M, & Concato J (2018). Suicidal ideation and behavior in US veterans with schizophrenia or bipolar disorder. Journal of Psychiatric Research, 102, 216–222. [DOI] [PubMed] [Google Scholar]
  25. Jahn DR, Bennett ME, Park SG, Gur RE, Horan WP, Kring AM, & Blanchard JJ (2016). The interactive effects of negative symptoms and social role functioning on suicide ideation in individuals with schizophrenia. Schizophrenia Research, 170(2), 271–277. 10.1016/j.schres.2015.12.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kelleher I, Corcoran P, Keeley H, Wigman JTW, Devlin N, Ramsay H, … Cannon M (2013). Psychotic Symptoms and Population Risk for Suicide Attempt: A Prospective Cohort Study. JAMA Psychiatry, 70(9), 940. 10.1001/jamapsychiatry.2013.140 [DOI] [PubMed] [Google Scholar]
  27. Kleiman EM, & Liu RT (2013). Social support as a protective factor in suicide: Findings from two nationally representative samples. Journal of Affective Disorders, 150(2), 540–545. 10.1016/j.jad.2013.01.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Klonsky ED, & May AM (2015). The Three-Step Theory (3ST): A New Theory of Suicide Rooted in the “Ideation-to-Action” Framework. International Journal of Cognitive Therapy, 8(2), 114–129. 10.1521/ijct.2015.8.2.114 [DOI] [Google Scholar]
  29. Kring AM, Gur RE, Blanchard JJ, Horan WP, & Reise SP (2013). The Clinical Assessment Interview for Negative Symptoms (CAINS): Final development and validation. Am J Psychiatry, 170(2), 165–172. (23377637). 10.1176/appi.ajp.2012.12010109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Loas G, Azi A, Noisette C, Legrand A, & Yon V (2009). Fourteen-year prospective follow-up study of positive and negative symptoms in chronic schizophrenic patients dying from suicide compared to other causes of death. Psychopathology, 42(3), 185–189. 10.1159/000209331 [DOI] [PubMed] [Google Scholar]
  31. Lopez-Morinigo J-D, Ayesa-Arriola R, Torres-Romano B, Fernandes AC, Shetty H, Broadbent M, … Dutta R (2016). Risk assessment and suicide by patients with schizophrenia in secondary mental healthcare: A case–control study. BMJ Open, 6(9), e011929. 10.1136/bmjopen-2016-011929 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Lopez-Morinigo J-D, Barrigón ML, Porras-Segovia A, Ruiz-Ruano VG, Martínez ASE, Escobedo-Aedo PJ, … Baca-García E (2021). Use of Ecological Momentary Assessment Through a Passive Smartphone-Based App (eB2) by Patients With Schizophrenia: Acceptability Study. Journal of Medical Internet Research, 23(7), e26548. 10.2196/26548 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Najas-Garcia A, Carmona VR, & Gómez-Benito J (2018). Trends in the Study of Motivation in Schizophrenia: A Bibliometric Analysis of Six Decades of Research. [DOI] [PMC free article] [PubMed]
  34. Parrish EM, Chalker SA, Cano M, Moore RC, Pinkham AE, Harvey PD, … Depp CA (2021). Ecological momentary assessment of interpersonal theory of suicide constructs in people experiencing psychotic symptoms. Journal of Psychiatric Research, 140, 496–503. 10.1016/j.jpsychires.2021.06.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Parrish EM, Depp CA, Moore RC, Harvey PD, Mikhael T, Holden J, … Granholm E (2020). Emotional determinants of life-space through GPS and ecological momentary assessment in schizophrenia: What gets people out of the house? Schizophrenia Research, 224, 67–73. 10.1016/j.schres.2020.10.002 [DOI] [PubMed] [Google Scholar]
  36. Porras-Segovia A, López-Moriñigo J, Álvarez R, Díaz-Oliván I, González V, & Baca-García E (2021). Psychiatric relapse following a phone app intervention: A case report. Psychiatry Research, 298, 113766. 10.1016/j.psychres.2021.113766 [DOI] [PubMed] [Google Scholar]
  37. Radomsky ED, Haas GL, Mann JJ, & Sweeney JA (1999). Suicidal Behavior in Patients With Schizophrenia and Other Psychotic Disorders. American Journal of Psychiatry, 156(10), 1590–1595. 10.1176/ajp.156.10.1590 [DOI] [PubMed] [Google Scholar]
  38. Reddy LF, Green MF, Rizzo S, Sugar CA, Blanchard JJ, Gur RE, … Horan WP (2014). Behavioral approach and avoidance in schizophrenia: An evaluation of motivational profiles. Schizophrenia Research, 159(1), 164–170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Reddy LF, Horan WP, & Green MF (2016). Motivational Deficits and Negative Symptoms in Schizophrenia: Concepts and Assessments. In Simpson EH & Balsam PD (Eds.), Behavioral Neuroscience of Motivation (pp. 357–373). Cham: Springer International Publishing. 10.1007/7854_2015_379 [DOI] [PubMed] [Google Scholar]
  40. Reinen J, Smith EE, Insel C, Kribs R, Shohamy D, Wager TD, & Jarskog LF (2014). Patients with schizophrenia are impaired when learning in the context of pursuing rewards. [DOI] [PubMed]
  41. Sedano-Capdevila A, Porras-Segovia A, Bello HJ, Baca-García E, & Barrigon ML (2021). Use of Ecological Momentary Assessment to Study Suicidal Thoughts and Behavior: A Systematic Review. Current Psychiatry Reports, 23(7), 41. 10.1007/s11920-021-01255-7 [DOI] [PubMed] [Google Scholar]
  42. Strauss GP, Waltz JA, & Gold JM (2013). A review of reward processing and motivational impairment in schizophrenia. Schizophrenia Bulletin, 40(Suppl_2), S107–S116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Tan M, Shallis A, & Barkus E (2020). Social anhedonia and social functioning: Loneliness as a mediator. PsyCh Journal, 9(2), 280–289. 10.1002/pchj.344 [DOI] [PubMed] [Google Scholar]
  44. Taylor CT, Pearlstein SL, & Stein MB (2020). A Tale of Two Systems: Testing a Positive and Negative Valence Systems Framework to Understand Social Disconnection Across Anxiety and Depressive Disorders. Journal of Affective Disorders, 266, 207–214. 10.1016/j.jad.2020.01.041 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Twisk JWR (2019). Mixed Model Analysis in Longitudinal Studies. In Practical Guides to Biostatistics and Epidemiology. Applied Mixed Model Analysis: A Practical Guide (2nd ed., pp. 131–150). Cambridge: Cambridge University Press. 10.1017/9781108635660.010 [DOI] [Google Scholar]
  46. Van Orden KA, Witte TK, Cukrowicz KC, Braithwaite SR, Selby EA, & Joiner TE Jr (2010). The interpersonal theory of suicide. Psychological Review, 117(2), 575. 10.1037/a0018697 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Villa J, Ehret BC, & Depp CA (2019). Systematic review of the inclusion of people with psychosis in suicide-specific clinical trials. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 41(3), 233–236. 10.1027/0227-5910/a000628 [DOI] [PubMed] [Google Scholar]
  48. Villa J, Pinkham AE, Kaufmann CN, Granholm E, Harvey PD, & Depp CA (2018). Interpersonal beliefs related to suicide and facial emotion processing in psychotic disorders. Journal of Psychiatric Research, 100, 107–112. 10.1016/j.jpsychires.2018.02.016 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES