Divergent semantic integration (DSI): Extracting creativity from narratives with distributional semantic modeling.

We developed a novel conceptualization of one component of creativity in narratives by integrating creativity theory and distributional semantics theory. We termed the new construct divergent semantic integration (DSI), defined as the extent to which a narrative connects divergent ideas. Across nine studies, 27 different narrative prompts, and over 3500 short narratives, we compared six models of DSI that varied in their computational architecture. The best-performing model employed Bidirectional Encoder Representations from Transformers (BERT), which generates context-dependent numerical representations of words (i.e., embeddings). BERT DSI scores demonstrated impressive predictive power, explaining up to 72% of the variance in human creativity ratings, even approaching human inter-rater reliability for some tasks. BERT DSI scores showed equivalently high predictive power for expert and nonexpert human ratings of creativity in narratives. Critically, DSI scores generalized across ethnicity and English language proficiency, including individuals identifying as Hispanic and L2 English speakers. The integration of creativity and distributional semantics theory has substantial potential to generate novel hypotheses about creativity and novel operationalizations of its underlying processes and components. To facilitate new discoveries across diverse disciplines, we provide a tutorial with code (osf.io/ath2s) on how to compute DSI and a web app ( osf.io/ath2s ) to freely retrieve DSI scores.

Automating creativity assessment with SemDis: An open platform for computing semantic distance.

Creativity research requires assessing the quality of ideas and products. In practice, conducting creativity research often involves asking several human raters to judge participants' responses to creativity tasks, such as judging the novelty of ideas from the alternate uses task (AUT). Although such subjective scoring methods have proved useful, they have two inherent limitations-labor cost (raters typically code thousands of responses) and subjectivity (raters vary on their perceptions and preferences)-raising classic psychometric threats to reliability and validity. We sought to address the limitations of subjective scoring by capitalizing on recent developments in automated scoring of verbal creativity via semantic distance, a computational method that uses natural language processing to quantify the semantic relatedness of texts. In five studies, we compare the top performing semantic models (e.g., GloVe, continuous bag of words) previously shown to have the highest correspondence to human relatedness judgements. We assessed these semantic models in relation to human creativity ratings from a canonical verbal creativity task (AUT; Studies 1-3) and novelty/creativity ratings from two word association tasks (Studies 4-5). We find that a latent semantic distance factor-comprised of the common variance from five semantic models-reliably and strongly predicts human creativity and novelty ratings across a range of creativity tasks. We also replicate an established experimental effect in the creativity literature (i.e., the serial order effect) and show that semantic distance correlates with other creativity measures, demonstrating convergent validity. We provide an open platform to efficiently compute semantic distance, including tutorials and documentation ( https://osf.io/gz4fc/ ).

Metacognition in argument generation: the misperceived relationship between emotional investment and argument quality.

Overestimation of one's ability to argue their position on socio-political issues may partially underlie the current climate of political extremism in the U.S. Yet very little is known about what factors influence overestimation in argumentation of socio-political issues. Across three experiments, emotional investment substantially increased participants' overestimation. Potential confounding factors like topic complexity and familiarity were ruled out as alternative explanations (Experiments 1-3). Belief-based cues were established as a mechanism underlying the relationship between emotional investment and overestimation in a measurement-of-mediation (Experiment 2) and manipulation-of-mediator (Experiment 3) design. Representing a new bias blind spot, participants believed emotional investment helps them argue better than it helps others (Experiments 2 and 3); where in reality emotional investment harmed or had no effect on argument quality. These studies highlight misguided beliefs about emotional investment as a factor underlying metacognitive miscalibration in the context of socio-political issues.

Validation of the Massachusetts Youth Screening Instrument with a looked after population.

The Massachusetts Youth Screening Instrument-version 2 (MAYSI-2) was developed to help identify mental health needs of young people admitted to youth detention centres. Only one study has applied the tool to a UK population and none have looked at young people who live in residential and secure care in Scotland. This study aimed to assess the validity of the MAYSI-2 in Scotland with a looked after and accommodated population. Boys and girls in a large education and care centre were asked to complete the MAYSI-2 within 72 hours of being accommodated. A total of 168 males and 69 females with a mean age of 15 completed the tool. Substantial levels of mental health need were identified. Girls appeared to have higher needs on all areas, bar alcohol and substance misuse. The MAYSI-2 had good internal consistency and exploratory factor analysis showed good overlap with the tool's original factor model. As a result, there can be more confidence in the validity and consistency of the tool with this population. This is also further evidence of the high need of this population, particularly girls.

Reliability and construct validity of the Automated Neuropsychological Assessment Metrics (ANAM) mood scale.

The reliability and construct validity of the Automated Neuropsychological Assessment Metrics (ANAM) mood scale (AMS) were examined using concurrent, well-validated measures of mood and confirmatory factor analysis (CFA) with a sample of 210 volunteer college students. The AMS was given in computerized format with multiple adjectives using a visual analog Likert scale yielding seven dimensions of mood including vigor, restlessness, depression, anger, fatigue, anxiety, and happiness. All seven mood dimensions of the AMS demonstrated excellent test-retest reliability and internal consistency. Also, the AMS anxiety dimension correlated strongly with the Spielberger's State Anxiety Inventory (r=0.67) and the AMS depression dimension correlated strongly with the Beck Depression Inventory-II (r=0.71). CFA revealed that the AMS 7-factor mood model fit the data well and significantly better than an alternative, theoretically plausible model. When concurrent measures of mood were incorporated in the CFA model, the AMS demonstrated both convergent and discriminant validity. The AMS 7-factor model explained 55.12% of the total variance in the items. It was concluded that the AMS provides a brief yet reasonably complete and valid assessment of mood.

Need for cognition as a predictor of psychosocial identity development.

The authors examined the hypothesis that psychosocial identity development is related to need for cognition (NFC), a social-cognitive individual-difference variable defined as the desire to engage in effortful thinking (J. T. Cacioppo, R. E. Petty, J. Feinstein, & W. Jarvis, 1996). They administered 2 measures of psychosocial identity-a scale from the Extended Objective Measure of Ego Identity Status 2 (EOMEIS-2; G. R. Adams, L. Bennion, & K. Huh, 1989) and the Identity subscale of the Erikson Psychosocial Stage Inventory (EPSI; D. A. Rosenthal, R. M. Gurney, & S. M. Moore, 1981)-and the NFC scale to 200 incoming college students and approximately half of those students about 15 months later. Results indicate that people with higher psychosocial identity levels had higher NFC scores at both time periods. In addition, higher Time 1 NFC scores were related to higher Time 2 EOMEIS-2 achieved scores and lower Time 2 foreclosure and diffusion scores, and changes in NFC over the course of the study were positively correlated with EPSI changes and negatively correlated with changes in EOMEIS-2 foreclosure and diffusion scores. Results provide support for the importance of a cognitive and motivational individual-difference variable in the development of a unique and cohesive identity.

Residential staff responses to adolescent self-harm: The helpful and unhelpful.

Adolescent self-harm is prevalent in residential and secure care and is the cause of distress to those harming themselves, to the staff caring for them and for other young people living with them. This article sought service user views on what staff supports were effective and what were counter-productive in order to improve the care offered to young people. Seven young people living in residential or secure care were interviewed. Thematic analysis was used to elicit key themes. Global themes of safety and care were elicited. The young people understood and accepted that the role of staff was to provide these. Within these themes, they noted numerous responses that had both helpful and unhelpful effects, including increased observation, removal of means and extra collaborative support. Service users made numerous recommendations to increase the helpful effects of staff support. Young people provided informed and helpful guidance on how best to care for them. Their views can help mental health professionals and care staff increase their helpful responses making them more effective and less counter-productive. This study is a rare representation of the views of young people in residential and secure care and how to respond to their self-harm behaviour.

Reflecting on explanatory ability: A mechanism for detecting gaps in causal knowledge.

People frequently overestimate their understanding-with a particularly large blind-spot for gaps in their causal knowledge. We introduce a metacognitive approach to reducing overestimation, termed reflecting on explanatory ability (REA), which is briefly thinking about how well one could explain something in a mechanistic, step-by-step, causally connected manner. Nine experiments demonstrated that engaging in REA just before estimating one's understanding substantially reduced overestimation. Moreover, REA reduced overestimation with nearly the same potency as generating full explanations, but did so 20 times faster (although only for high complexity objects). REA substantially reduced overestimation by inducing participants to quickly evaluate an object's inherent causal complexity (Experiments 4-7). REA reduced overestimation by also fostering step-by-step, causally connected processing (Experiments 2 and 3). Alternative explanations for REA's effects were ruled out including a general conservatism account (Experiments 4 and 5) and a covert explanation account (Experiment 8). REA's overestimation-reduction effect generalized beyond objects (Experiments 1-8) to sociopolitical policies (Experiment 9). REA efficiently detects gaps in our causal knowledge with implications for improving self-directed learning, enhancing self-insight into vocational and academic abilities, and even reducing extremist attitudes.

Detecting subtle expressions: older adults demonstrate automatic and controlled positive response bias in emotional perception.

The present study examined age differences in emotional perception for the detection of low-intensity, single-emotion facial expressions. Confirming the "positivity effect," at 60 ms and 2,000 ms presentation rates older adults (age = 61+ years, n = 39) exhibited a response bias favoring happy over neutral responses, whereas younger adults (age = 18-23 years, n = 40) favored neutral responses. Furthermore, older adults favored neutral over fearful responses at the 60 ms presentation rate, relative to younger adults. The finding that age differences in response bias were most pronounced at the 60 ms versus 2,000 ms presentation rate suggests that positivity effects in emotional perception rely partly on automatic processing.

Initial clinical validation of an embedded performance validity measure within the automated neuropsychological metrics (ANAM).

The measurement of effort and performance validity is essential for computerized testing where less direct supervision is needed. The clinical validation of an Automated Neuropsychological Metrics-Performance Validity Index (ANAM-PVI) was examined by converting ANAM test scores into a common metric based on their relative infrequency in an outpatient clinic sample with presumed good effort. Optimal ANAM-PVI cut-points were determined using receiver operator characteristic (ROC) curve analyses and an a priori specificity of 90%. Sensitivity/specificity was examined in available validation samples (controls, simulators, and neurorehabilitation patients). ANAM-PVI scores differed between groups with simulators scoring the highest. ROC curve analysis indicated excellent discriminability of ANAM-PVI scores ≥5 to detect simulators versus controls (area under the curve = 0.858; odds ratio for detecting suboptimal performance = 15.6), but resulted in a 27% false-positive rate in the clinical sample. When specificity in the clinical sample was set at 90%, sensitivity decreased (68%), but was consistent with other embedded effort measures. Results support the ANAM-PVI as an embedded effort measure and demonstrate the value of sample-specific cut-points in groups with cognitive impairment. Examination of different cut-points indicates that clinicians should choose sample-specific cut-points based on sensitivity and specificity rates that are most appropriate for their patient population with higher cut-points for those expected to have severe cognitive impairment (e.g., dementia or severe acquired brain injury).

Emotional attention set-shifting and its relationship to anxiety and emotion regulation.

Attentional deployment is a primary strategy individuals use to regulate emotion. In 2 experiments, a measure of an individual's ability to deploy attention toward and away from emotional mental representations was developed. This measure of attentional control capacity for emotion adapted an explicit-cuing task switching paradigm in which participants had to shift between emotional and neutral mental sets. Experiment 1 (N = 118) showed that those higher in trait anxiety and worrisome thoughts took longer to switch from a neutral to an emotional mental set. In Experiment 2 (N = 42), participants were given a stressful anagram task, and those who switched more efficiently from a neutral set to an emotional set were more frustrated by the stressful task. In addition, those who switched more efficiently from an emotional set to a neutral set persisted longer on the stressful task. These findings provide an initial step toward identifying possible mechanisms through which individuals apply attentional control to emotional mental representations to regulate emotion.

Individuals lower in working memory capacity are particularly vulnerable to anxiety's disruptive effect on performance.

Anxiety has a disruptive effect on performance in a number of domains. The purpose of this study was to determine whether individual differences in working memory (WM) capacity are related to an individual's susceptibility to anxiety's detrimental effect on performance. Fifty undergraduate students (28 females) were administered the State-Trait Anxiety Inventory (STAI) (Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983) to measure trait anxiety and the automated operation span task to measure WM capacity (Unsworth, Heitz, Schrock, & Engle, 2005). Then, they performed a highly demanding dual-task that consisted of a primary short-term memory task and a secondary tone-discrimination task that served as a measure of spare capacity. Anxiety and WM capacity interacted to affect performance on the auditory task so that those low in WM capacity were particularly vulnerable to anxiety's disruptive effect, whereas those high in WM capacity were buffered against anxiety's effect. These findings suggest that WM capacity may be an important factor in determining which individuals underperform on anxiety-provoking tests such as scholastic achievement tests.

Abstract: A Multi-Level Model of Individual Differences in Speed/Accuracy Tradeoff (SATin).

Individuals performing an experimental cognitive task have a choice whether to favor accuracy, speed, or weight them both equally. Models of speed/ accuracy tradeoff have been proposed in the assessment literature ( van der Linden, 2007 ) and experimental literature ( Ratcliff & Rouder, 1998 ). However, these models do not estimate individual differences in choice of speed/ accuracy tradeoff at between- and within-subjects levels. The top of Figure 1 presents the equations and path diagram for the SATin model. Individual differences in speed/ accuracy tradeoff will be modeled at two levels with, 1) variability in Tradeoff (between-subject level, Level 2) and 2) variability in c (within-subject level, Level 1). An individual's Tradeoff factor score represents the individual's distributional position relative to others regarding whether they favor speed (values < 0), accuracy (values > 0), or neither (value = 0). A negative c indicates that the individual is trading off speed and accuracy for these particular trials, whereas a positive and zero c indicate the individual is not trading off. [Figure: see text] Panels 1 and 2 illustrate Simpson's paradox, where speed/ accuracy tradeoff occurs in opposing directions at between- and within-subject levels. This highlights the need for a multi-level model as the researcher would draw opposing conclusions by observing only one of these levels. Simulations studies compared the SATin model to a popular model in cognitive psychology that uses speed alone to estimate ability. SATin outperformed this model by accounting for substantially more variance in actual ability.