Examining the item response process to personality measures in high-stakes situations: Issues of measurement validity and predictive validity

We conducted two experimental studies with between-subjects and within-subjects designs to investigate the item response process for personality measures administered in high- versus low-stakes situations. Apart from assessing measurement validity of the item response process, we examined predictive validity; that is, whether or not different response models entail differential selection outcomes. We found that ideal point response models fit slightly better than dominance response models across high- versus low-stakes situations in both studies. Additionally, fitting ideal point models to the data led to fewer items displaying differential item functioning compared to fitting dominance models. We also identified several items that functioned as intermediate items in both the faking and honest conditions when ideal point models were fitted, suggesting that ideal point model is “theoretically” more suitable across these contexts for personality inventories. However, the use of different response models (dominance vs. ideal point) did not have any substantial impact on the validity of personality measures in high-stakes situations, or the effectiveness of selection decisions such as mean performance or percent of fakers selected. These findings are significant in that although prior research supports the importance and use of ideal point models for measuring personality, we find that in the case of personality faking, though ideal point models seem to have slightly better measurement validity, the use of dominance models may be adequate with no loss to predictive validity.     

The Value of Perception

This paper develops a form of transcendental naïve realism. According to naïve realism, veridical perceptual experiences are essentially relational. According to transcendental naïve realism, the naïve realist theory of perception is not just one theory of perception amongst others, to be established as an inference to the best explanation and assessed on the basis of a cost-benefit analysis that weighs performance along a number of different dimensions: for instance, fidelity to appearances, simplicity, systematicity, fit with scientific theories, and so on. Rather, naïve realism enjoys a special status in debates in the philosophy of perception because it represents part of the transcendental project of explaining how it is possible that perceptual experience has the distinctive characteristics it does. One of the potentially most interesting prospects of adopting a transcendental attitude towards naïve realism is that it promises to make the naïve realist theory of perception, in some sense, immune to falsification. This paper develops a modest form of transcendental naïve realism modelled loosely on the account of the reactive attitudes provided by Strawson in ‘Freedom and Resentment’, and suggests one way of understanding the claim that naïve realism is immune to falsification.     

Naloxone administration impairs autoshaped learning

Effects of naloxone on acquisition of autoshaped behavior were investigated. Rats deprived to 85% of free-feeding weights were trained to touch a retractable lever; delivery of a food pellet occurred on every trial following lever retraction. The lever was retracted immediately if a touch occurred within 15 s, or automatically after 15 s. Analyses were conducted on number and latencies of touches of the extended lever, nose-pokes (touches) directed at the retracted lever during intertrial intervals (a measure less constrained by ceiling effects than extended lever touches), and unconditioned exploratory rearing activity, measured as touches of a metal strip mounted above the grid floor of the apparatus. In an initial experiment, male Sprague-Dawley rats were given saline or naloxone (2.0 mg/kg, ip) 5 min before a training session of 12 trials. Two days later they were tested, in the absence of drug, in a session of 36 (three blocks of 12) trials. Naloxone depressed training levels of lever responding, in addition to slowing acquisition rate. No effect of naloxone was observed on rearing activity. Previous work showed that injection of saline 5 min before behavioral testing increases the rate of autoshaping compared to injections 30 min before (Messing & Sparber, 1984). Thus, effects of naloxone on acquisition of lever-directed behaviors may have been confounded by behavioral depressant effects and/or by an injection effect such a short time before testing. In a second experiment naloxone (0.5 or 2.0 mg/kg) was injected after five of seven training sessions (12 trials each) to male and female rats. A 6-s delay of reinforcement was inserted between lever retraction and food delivery, slowing acquisition rates and providing the opportunity to test the effects of naloxone throughout a multiple-session task. The low dose retarded acquisition of extended lever touching in both sexes; both doses retarded acquisition of interim lever touching in males. Thus, in some circumstances, post-training naloxone administration may impair learning. The results support the notion that low doses of naloxone may have agonist activity.     

The typology of parricide and the role of mental illness: Data-driven approach

Parricide is a rare type of homicide in which mental illness is often an important factor. The aims of this study were (a) to describe the characteristics of parricide offenders with a focus on mental illness and clinical care and (b) to examine Heide's widely used typology of parricide through a data-driven approach. We analyzed all homicides in England and Wales between 1997 and 2014. Parricide offenders in our sample were most often male, unmarried, and unemployed, with a third of offenders diagnosed with schizophrenia; 28% had been in contact with mental health services before the offense. The latent class analysis resulted in three types of parricide offenders: middle-aged with affective disorder, previously abused, and seriously mentally Ill, which confirmed, to an extent, Heide's typology. Health and social care services should actively engage with carers of people with mental illness and support to those caring for older relatives and victims of abuse.     

Psychometric and Validity Issues in Machine Learning Approaches to Personality Assessment: A Focus on Social Media Text Mining

In the age of big data, substantial research is now moving toward using digital footprints like social media text data to assess personality. Nevertheless, there are concerns and questions regarding the psychometric and validity evidence of such approaches. We seek to address this issue by focusing on social media text data and (i) conducting a review of psychometric validation efforts in social media text mining (SMTM) for personality assessment and discussing additional work that needs to be done; (ii) considering additional validity issues from the standpoint of reference (i.e. ‘ground truth’) and causality (i.e. how personality determines variations in scores derived from SMTM); and (iii) discussing the unique issues of generalizability when validating SMTM for personality assessment across different social media platforms and populations. In doing so, we explicate the key validity and validation issues that need to be considered as a field to advance SMTM for personality assessment, and, more generally, machine learning personality assessment methods. © 2020 European Association of Personality Psychology     

Selective exposure and dissonance after decisions

Well-known literature reviews from the 1960s question whether cognitive dissonance underlies experimental participants' selective exposure of themselves to consonant messages and avoidance of dissonant ones. A meta-analytic review of 16 studies published from 1956 to 1996 and involving 1,922 total participants shows that experimental tests consistently support the supposition that dissonance is associated with selective exposure (r = .22, p < .001). Statistical power exceeded .99. Advances in statistical methodology and increased attention to selecting appropriate tests of dissonance theory were essential to finally resolving this question.     

Perception and production of rise-fall intonation in American English

At the segmental level, the rate of speaking affects the degree of physical undershoot of articulatory targets and the resulting perception. Little is known regarding evidence of these effects at the suprasegmental level, particularly in intonation. In this study, the effect of rate of speaking on fundamental frequency and on perceptual judgments of peak pitch in a rise-fall intonation pattern was investigated. First, speakers produced rise-fall intonations in sentence contexts at slow, normal, and fast speaking rates. Peak fundamental frequencies (F0) of the slow productions were significantly lower than those of the normal or fast productions. The mean normal rate production of the word Miami was used as a model for the target word in a series of subsequent perceptual experiments. Altering the duration of the target word to represent slow, normal, and fast rates of speaking did not affect listener judgment of peak pitch. Finally, the pitch of the target word was measured in a sentence context. No differences between peak pitch in isolation or in sentence context were found. It was concluded that the production and perception of this form of intonation was not subject to the effects of rate that are seen at the segmental level.     

Specificity of the MMPI-2 Fake Bad Scale as a marker for personal injury malingering

Psychologists who evaluate patients in medicolegal contexts should utilize objective assessment data with empirically established sensitivity and specificity for identifying negative response bias. The purpose of this study was to investigate the specificity of the Fake Bad Scale for identifying negative response bias in personal injury claimants. The cutoff scores proposed by Lees-Haley and colleagues were applied a federal prison, medical outpatients, and patients from to inmate volunteers from substance abuse unit. Half of the inmates were given instructions to malinger psychopathology to affect the adjudication process, and the remaining inmates and all of the hospital patients were given standard instructions. The original cutoff scores correctly identified the majority of inmates instructed to malinger psychopathology, but these scores resulted in unacceptably high rates of false positive classifications. The revised cutoff scores resulted in fewer false positives, i.e., 8%-24%.