Conceptual complexity and the bias/variance tradeoff

In this paper we propose that the conventional dichotomy between exemplar-based and prototype-based models of concept learning is helpfully viewed as an instance of what is known in the statistical learning literature as the bias/variance tradeoff. The bias/variance tradeoff can be thought of as a sliding scale that modulates how closely any learning procedure adheres to its training data. At one end of the scale (high variance), models can entertain very complex hypotheses, allowing them to fit a wide variety of data very closely—but as a result can generalize poorly, a phenomenon called overfitting. At the other end of the scale (high bias), models make relatively simple and inflexible assumptions, and as a result may fit the data poorly, called underfitting. Exemplar and prototype models of category formation are at opposite ends of this scale: prototype models are highly biased, in that they assume a simple, standard conceptual form (the prototype), while exemplar models have very little bias but high variance, allowing them to fit virtually any combination of training data. We investigated human learners’ position on this spectrum by confronting them with category structures at variable levels of intrinsic complexity, ranging from simple prototype-like categories to much more complex multimodal ones. The results show that human learners adopt an intermediate point on the bias/variance continuum, inconsistent with either of the poles occupied by most conventional approaches. We present a simple model that adjusts (regularizes) the complexity of its hypotheses in order to suit the training data, which fits the experimental data better than representative exemplar and prototype models.     

The effect of stopping before turning on the direct observational measure of whole body turning bias

Turning bias, the preferential tendency to turn toward a given direction has been reported in both rodents and human participants. The observational gait method of determining turning bias in humans requires a stop prior to turning. This study removed the stop and hypothesised that turning bias would remain the same between stop and non-stop conditions if bias was solely under the control of neurochemical asymmetries. The results showed that statistically turning bias remained the same (to the left) regardless of method used but there was no agreement between the methods thus rejecting the hypothesis. It is likely that when not stopping biomechanical factors related to gait when turning influence the direction of turn rather than solely neurochemical asymmetries.     

Cognitive inhibition in depression

Cognitive inhibition is a key mechanism in the regulation of emotion. There is emerging evidence that depression is characterized by deficits in the inhibition of mood-congruent material. These deficits could result in prolonged processing of negative, goal-irrelevant aspects of presented information thereby hindering recovery from negative mood and leading to the sustained negative affect that characterizes depressive episodes. Indeed, it has been suggested that deficits in cognitive inhibition lie at the heart of memory and attention biases in depression, and set the stage for ruminative responses to negative events and negative mood states. A ruminative response style results in a heightened vulnerability to experience episodes of major depression. Recent research has demonstrated that deficient inhibition of negative material is associated with heightened rumination. In this article, we review the depression literature with a focus on studies that investigate cognitive inhibition in depressed participants and in participants who report a history of major depressive episodes. In addition, we summarize neurobiological findings that indicate a strong relation between depression and deficits in inhibition and we take a closer look at the relation of inhibition, rumination and mood regulation.     

An integrated model of social psychological and personality psychological perspectives on personality and wellbeing

This article uses multi-rater data from 458 triads (students, mother, father, total N = 1374) to examine the relationship of personality ratings with wellbeing ratings, using a multi-method approach to separate accurate perceptions (shared across raters) from biased perceptions of the self (rater-specific variance). The social-psychological perspective predicts effects of halo bias in self-ratings on wellbeing, whereas the personality-psychological perspective predicts effects of personality traits on wellbeing. Results are more consistent with the personality perspective in that neuroticism (negative), extraversion, agreeableness, and to a lesser extent conscientiousness predicted wellbeing, whereas positive illusions about the self were only weakly and not significantly related to wellbeing. These results cast doubt on the hypothesis that self-enhancement is beneficial for wellbeing.     

Base Rates,Deception Detection,and Deception Theory: A Reply to Burgoon (2015)

This essay continues a debate about the relative scientific merits of the Park‐Levine Model (PLM) and Interpersonal Deception Theory (IDT) with regard to accuracy in deception detection. Key points of disagreement include (a) the degree to which message recipients are sensitive to sender veracity and (b) the extent to which interactivity moderates the veracity and base‐rate effects specified by PLM. According to PLM, people are truth‐biased, truth‐bias makes people insensitive to deception, and as a consequence, base rates affect accuracy regardless of interactivity. IDT, in direct contrast, holds that people are sensitive to the veracity of other's communication and that interactivity is a key moderator. Consistent with PLM, data repeatedly show that people are insensitive to other's veracity and that this insensitivity is general across media affordances including interactivity. This rejoinder uses Burgoon's (2015) own interactive data to demonstrate the empirical superiority of PLM over IDT.     

Stereotype accuracy

Methodological issues and empirical findings from stereotype accuracy research are reviewed. Methodological issues include the limitations of accuracy criteria and three methods of comparing perceived to actual group characteristics--signed discrepancies, absolute discrepancies, and within-subject correlations. Empirical findings concern the roles of cognitive processes, status and power, and social ideology in intergroup perceptions; stereotype development; individual differences in stereotyping; and stereotype use. It is argued that stereotype accuracy research is neither easy to do nor politically popular, but that it may challenge existing theory and stimulate new ideas about the nature of stereotyping processes.     

Interpersonal self‐support and attentional bias on negative and positive interpersonal information

Interpersonal self‐support is an indigenous Chinese personality concept. It represents the idealized notion of the kind of personality traits that help individuals deal with interpersonal problems and develop and maintain the harmonic and appropriate social relationships required in China's collectivistic and interdependent culture. It also was assumed to be a protective personality factor with regard to mental health and was found to be negatively related to psychosomatic symptoms. In the current study, cognitive processing of interpersonal information is assumed to be an underlying mechanism that connects interpersonal self‐support with interpersonal relationships and mental health. To test this hypothesis, we conducted two experiments to investigate whether attentional bias on positive and negative interpersonal information was related to high and low interpersonal self‐support. A spatial cueing task and the emotional Stroop task were administered to two samples of high and low interpersonal self‐support Chinese undergraduate students to measure attentional bias. The results from both experiments suggested that high interpersonal self‐support students had an attentional bias toward positive interpersonal information, while low interpersonal self‐support students preferentially attended to negative interpersonal information. Study 1 indicated that attentional bias toward positive interpersonal information was easily engaged in the high interpersonal self‐support group, while attentional bias toward negative interpersonal information was both easy to engage and difficult to disengage in the low interpersonal self‐support students. These results support our hypotheses that high interpersonal self‐support people engage in positive processing of interpersonal information, whereas low interpersonal self‐support people engage in negative processing of interpersonal information. The differential balance between positive and negative processing on interpersonal information may explain why interpersonal self‐support predicts both mental health and interpersonal relationships. In addition, the relational schema may explain why interpersonal self‐support is associated with an attentional bias toward interpersonal information.     

Does mindfulness attenuate thoughts emphasizing negativity,but not positivity?

The current research investigated whether mindfulness is differentially associated with thoughts that emphasize positive or negative valence. In Study 1, trait mindfulness was inversely associated with negative rumination but unassociated with positive rumination, controlling for state affect. In Study 2, participants completed either a mindful breathing meditation or a comparable control exercise, followed by a thought listing while viewing affective images. Compared to the control condition, the mindfulness condition listed proportionately fewer negative thoughts, particularly in response to negative images, and more non-valenced thoughts. The conditions did not differ in their proportions of positive thoughts. These results suggest that mindfulness may attenuate thoughts that emphasize negativity but not those that emphasize positivity.     

Perceived Control and the Optimistic Bias: A Meta-Analytic Review

People consistently believe that negative events are less likely to happen to them than to others. Research suggests a relationship between this optimistic bias and perceived control such that the greater control people perceive over future events, the greater their optimistic bias. We conducted a meta-analysis of 27 independent samples to quantify the size of this relationship and examine what variables moderated the relationship. Greater perceived control was significantly related to greater optimistic bias, but this relationship was moderated by participant nationality, student status, risk status, and the type of optimistic bias measure used. We discuss the findings in the context of primary versus secondary control.