Extracting the Variance Inflation Factor and Other Multicollinearity Diagnostics from Typical Regression Results

Multicollinearity is a potential problem in all regression analyses. However, the examination of multicollinearity is rarely reported in primary studies. In this article we discuss and show several post hoc methods for assessing multicollinearity. One such multicollinearity diagnostic is the variance inflation factor. We outline the post hoc variance inflation factor method, which computes the variance inflation factor from the standardized regression coefficient and semi-partial correlation, both of which can be calculated from commonly reported regression results. Three examples of computing multicollinearity diagnostics using data from published studies are shown. We conclude with a discussion and practical implications.     

Cognitive fusion and emotion differentiation: does getting entangled with our thoughts dysregulate the generation,experience and regulation of emotion?

We tested whether cognitive fusion impairs emotion differentiation and thereby mediates relations between cognitive fusion and depression and panic symptoms among 55 adults (M_age?=?26.8 (3.9), 50.9% women). Using visual stimuli, we elicited multiple emotion states and measured (a) emotional intensity – the subjective emotion intensity of elicited emotions (i.e. Specific Emotion Intensity – SEI), as well as (b) emotional differentiation – the degree of co-activation of multiple negative emotions when a specific emotion was elicited (i.e. Multiple Emotion Co-Activation – MECA). First, as hypothesised, we found that cognitive fusion predicted lower levels of emotion differentiation (MECA). In contrast, as hypothesised, these effects were significantly greater than the (null) effects of cognitive fusion on emotion intensity (SEI). Second, as predicted, MECA, but not SEI, predicted depression and panic symptoms. Finally, we found that MECA mediated the effects of cognitive fusion on depression and panic symptoms. The present findings contribute novel, preliminary empirical insight into associations between cognitive fusion, impaired emotion differentiation and mental ill-health.     

The human Turing machine: a neural framework for mental programs

In recent years much has been learned about how a single computational processing step is implemented in the brain. By contrast, we still have surprisingly little knowledge of the neuronal mechanisms by which multiple such operations are sequentially assembled into mental algorithms. We outline a theory of how individual neural processing steps might be combined into serial programs. We propose a hybrid neuronal device: each step involves massively parallel computation that feeds a slow and serial production system. Production selection is mediated by a system of competing accumulator neurons that extends the role of these neurons beyond the selection of a motor action. Productions change the state of sensory and mnemonic neurons and iteration of such cycles provides a basis for mental programs.     

Religiosity and social welfare: competing influences of cultural conservatism and prosocial value orientation

This research examines the hypothesis that religiosity has two competing psychological influences on the social welfare attitudes of contemporary Americans. On the one hand, religiosity promotes a culturally based conservative identity, which in turn promotes opposition to federal social welfare provision. On the other hand, religiosity promotes a prosocial value orientation, which in turn promotes support of federal social welfare provision. Across two national samples (Ns = 1,513 and 320) and one sample of business employees (N = 710), reliable support for this competing pathways model was obtained. We argue that research testing influences of nonpolitical individual differences on political preferences should consider the possibility of competing influences that are rooted in a combination of personality processes and contextual-discursive surroundings.     

Genetic and environmental influences on girls' and boys' gender-typed and gender-neutral values

In this first investigation of genetic and environmental influences on children's values, 271 German twin pairs (50.2% boys) reported their values at ages 7-11 years using the Portrait Values Questionnaire (Schwartz & Rubel, 2005). We distinguished between gender-neutral (conservation vs. openness to change) and gender-typed (self-transcendence vs. self-enhancement) values. Boys differed from girls in the importance given to gender-typed benevolence, achievement, and power values. Gender-neutral values showed moderate (.34) and gender-typed values showed higher (.49) heritability, with nonshared environment and error accounting for the remaining variance. For both sexes, substantial genetic effects accounted for the importance children gave to their respective gender-stereotypical end of the self-transcendence versus self-enhancement dimension. However, dramatic sex differences emerged in the gender-atypical end of the distribution. For girls, low self-transcendence (high gender-atypical values) showed a large (.76) group heritability. For boys, gender-atypical values (high self-transcendence) showed no heritability and a modest (.10) shared environment effect.     

Stability and change in intelligence from age 11 to ages 70, 79, and 87: the Lothian Birth Cohorts of 1921 and 1936

Investigating the predictors of age-related cognitive change is a research priority. However, it is first necessary to discover the long-term stability of measures of cognitive ability because prior cognitive ability level might contribute to the amount of cognitive change experienced within old age. These two issues were examined in the Lothian Birth Cohorts of 1921 and 1936. Cognitive ability data were available from age 11 years when the participants completed the Moray House Test No. 12 (MHT). The Lothian Birth Cohort 1936 (LBC1936) completed the MHT a second time at age 70. The Lothian Birth Cohort 1921 (LBC1921) completed the MHT at ages 79 and 87. We examined cognitive stability and change from childhood to old age in both cohorts, and within old age in the LBC1921. Raw stability coefficients for the MHT from 11-70, 11-79, and 11-87 years were .67, .66, and .51, respectively; and larger when corrected for range restriction in the samples. Therefore, minimum estimates of the variance in later-life MHT accounted for by childhood performance on the same test ranged from 26-44%. This study also examined, in the LBC1921, whether MHT score at age 11 influenced the amount of change in MHT between ages 79 and 87. It did not. Higher intelligence from early life was apparently protective of intelligence in old age due to the stability of cognitive function across the lifespan, rather than because it slowed the decline experienced in later life.     

Sorry is the Hardest Word to Say: The Role of Self-Control in Apologizing

Apologizing is an effective strategy for reconciling relationships after transgressions. However, transgressors often resist or refuse to apologize. The current research investigated the role of self-control in apologizing. In Study 1, self-control was associated with participants’ proclivity to apologize and apologetic and nonapologetic behavior. In Studies 2 and 3, self-control was manipulated to test the causal relationship. Both studies found participants with high self-control were more apologetic and less nonapologetic and were more likely to use apologetic statements in e-mails to their victims. Overall, these studies suggest that transgressors with high self-control are more apologetic than those with low self-control.     

Gender differences in egalitarian behavior and attitudes in early childhood

It is axiomatic that human females are more egalitarian than males. However, surprisingly little empirical research exists that empirically assesses this stereotype. We created two studies designed to objectively examine egalitarian behavior in 3‐ to 5‐year‐old children. In Study 1 we compared the egalitarian behavior and attitudes of American girls versus boys by tabulating the extent to which each gender awarded the same number of stickers to, and liked to the same degree, two puppets. One puppet followed the child's instructions or actions while the other did not during a drawing task in which the child played the roles of leader and peer. In the peer role, girls exhibited more egalitarian behavior than boys. In Study 2, French‐Canadian children were shown two drawings by unknown peers—one messily and one neatly colored—, then asked to distribute stickers to each peer's drawing. Again, girls exhibited more egalitarian behavior than boys. Results suggest the origins of gender differences in egalitarian behavior occur early in life and merit further investigation.     

Presleep attributions about arousal during sleep: nocturnal panic

This study examined the role of presleep attributions about physiological events during sleep in nocturnal panic attacks. Patients who regularly experienced nocturnal panic were physiologically monitored as audio signals were presented during sleep. They were randomly assigned to 1 of 3 conditions: expected, in which signals of intense physiological changes were expected; unexpected, in which signals of intense physiological changes were not expected; or control, involving distinctly different signals unrelated to physiological responses. The unexpected condition led to substantially more self-reported distress and panic attacks. The experimental conditions did not elicit different autonomic reactions, but those who panicked showed stronger physiological responses than those who did not panic. The findings are consistent with a cognitive model of nocturnal panic attacks.     

Is the processing of words during eye fixations in reading strictly serial?

Extant models of oculomotor control during reading maintain that allocation of attention confines word recognition to one (target) word at a time, and that an eye movement to a new (posttarget) word is computed before attention is shifted to it. To test these assumptions, properties of the posttarget's preview were manipulated during the fixation of the preceding target word. The main results revealed longer target viewing durations when the posttarget preview was visually distinctive or when it was orthographically illegal. The meaning of posttarget text did not affect initial target word reading, although it affected the time spent rereading the target. To account for these findings, extant attention-shift models must assume that readers obtain at least visuospatial and orthographic information from a parafovelly visible word before it is attended. This view has shortcomings, however, and several considerations favor less restrictive model assumptions according to which attention can be allocated to more than one word at a time.