Single-Stage Prediction Models Do Not Explain the Magnitude of Syntactic Disambiguation Difficulty

The disambiguation of a syntactically ambiguous sentence in favor of a less preferred parse can lead to slower reading at the disambiguation point. This phenomenon, referred to as a garden-path effect, has motivated models in which readers initially maintain only a subset of the possible parses of the sentence, and subsequently require time-consuming reanalysis to reconstruct a discarded parse. A more recent proposal argues that the garden-path effect can be reduced to surprisal arising in a fully parallel parser: words consistent with the initially dispreferred but ultimately correct parse are simply less predictable than those consistent with the incorrect parse. Since predictability has pervasive effects in reading far beyond garden-path sentences, this account, which dispenses with reanalysis mechanisms, is more parsimonious. Crucially, it predicts a linear effect of surprisal: the garden-path effect is expected to be proportional to the difference in word surprisal between the ultimately correct and ultimately incorrect interpretations. To test this prediction, we used recurrent neural network language models to estimate word-by-word surprisal for three temporarily ambiguous constructions. We then estimated the slowdown attributed to each bit of surprisal from human self-paced reading times, and used that quantity to predict syntactic disambiguation difficulty. Surprisal successfully predicted the existence of garden-path effects, but drastically underpredicted their magnitude, and failed to predict their relative severity across constructions. We conclude that a full explanation of syntactic disambiguation difficulty may require recovery mechanisms beyond predictability.     

The Greta Thunberg Effect: Familiarity with Greta Thunberg predicts intentions to engage in climate activism in the United States

Despite Greta Thunberg's popularity, research has yet to investigate her impact on the public's willingness to take collective action on climate change. Using cross-sectional data from a nationally representative survey of U.S. adults (N = 1,303), we investigate the “Greta Thunberg Effect,” or whether exposure to Greta Thunberg predicts collective efficacy and intentions to engage in collective action. We find that those who are more familiar with Greta Thunberg have higher intentions of taking collective actions to reduce global warming and that stronger collective efficacy beliefs mediate this relationship. This association between familiarity with Greta Thunberg, collective efficacy beliefs, and collective action intentions is present even after accounting for respondents’ overall support for climate activism. Moderated mediation models testing age and political ideology as moderators of the “Greta Thunberg Effect” indicate that although the indirect effect of familiarity with Greta Thunberg via collective efficacy is present across all age-groups, and across the political spectrum, it may be stronger among those who identify as more liberal (than conservative). Our findings suggest that young public figures like Greta Thunberg may motivate collective action across the U.S. public, but their effect may be stronger among those with a shared political ideology. Implications for future research and for broadening climate activists’ appeals across the political spectrum are discussed.     

Some recommendations for doing high‐impact research in social psychological science

In this article, I reflect upon my own career to distil some general recommendations for doing high‐impact social psychological research in the midst of a worldwide pandemic. My suggestions include (a) testing and deriving theories that can help explain real‐world human judgment and behaviour (answering questions that people care about), (b) preaching beyond the choir (communicating your research to audiences outside of social psychology), and (c) birds of a feather are stronger together (maximizing impact through strategic collaborations).     

Action representation deficits in adolescents with developmental dyslexia

Developmental dyslexia (DD), a severe and frequent disorder of reading acquisition, is characterized by a diversity of cognitive and motor deficits whose interactions still remain under debate. Although deficits in the automatization of sensorimotor control have been highlighted, internal action representation allowing prediction has never before been investigated. In this study, we considered action representation of 18 adolescents with pure DD and 18 age-matched typical readers. Participants actually and mentally performed a visually guided pointing task involving strong spatiotemporal constraints (speed/accuracy trade-off paradigm). While actual and mental movement times of typical readers were isochronous and both conformed to Fitts’ law, the movement times of dyslexics differed between conditions, and only the actual movement times conformed to Fitts’ law. Furthermore, the quality of motor imagery correlated with word reading abilities. This suggests that the process of action representation is impaired in pure DD and supports the sensorimotor perspective of DD. Theoretical implications are discussed.     

Temperamental Factors in the Execution of Interrupted Editing Tasks

The question investigated in this study is how the temperament traits of strength of excitation (SE), strength of inhibition (SI), and mobility (MO) affect behaviour during the execution of computer tasks and tasks interrupting them. Several hypotheses, partly derived from the regulative theory of temperament, were tested in an experiment in which the natural environment and the types of task commonly performed by secretaries were simulated. The sample consisted of 39 female secretaries (21–64 years old). It was found that individuals high on SE and MO needed less time to resume tasks after an interruption than individuals low on these temperament traits. Interruptions similar to the main task resulted in longer resumption times for low-SE individuals. MO was negatively related to the time needed to perform the main task. A model describing the relationship between the frequency of switches between tasks and the speed of task performance in low-MO and high-MO individuals is presented.     

Anger and aggression in women: Influence of sports choice and testosterone administration

We report on two studies of anger and aggression in women. One study concerns an experimental study of anger induction in aggressive and non-aggressive sportswomen. It was found that sports choice in itself, contrary to expectation, does not predict anger arousal and aggressive behavior in the laboratory. However, at an individual level the anger proneness of the subject, as measured by a questionnaire we developed, was related to the intensity of aggressive behavior and subjectively reported anger. The second study concerns the activating effects of androgens on aggression and anger proneness. In a group of 22 female-to-male transsexuals, a battery of anger proneness and aggression questionnaires was administered twice: shortly before and 3 months after the start of androgen treatment. Administration of androgens was clearly associated with an increaese in anger proneness, although there were no changes in several aspects of overt aggressive behavior. © 1994 Wiley-Liss, Inc.     

Perceptions of War Trauma and Healing of Marital Relations Among Torture-surviving Congolese Couples Participating in Multicouple Therapy

Citizens of the Democratic Republic of Congo (DRC) experienced widespread torture during national wars between 1998 and 2003. Couples who survived and stayed intact suffered tremendous relationship stress. This study used a critical ethnography framework to explore the prewar, wartime, and postwar experiences of 13 torture-surviving couples who participated in a 10-session Torture-Surviving Couple Group in 2008 in the DRC. The group was designed to address the relational effects of torture and war trauma. Participants reported profound negative effects of the war on their relationships; mostly positive experiences during the group, including marital and peer connection and relationship growth; and a number of improvements in their relationship after the group. Implications include support for the use of relational interventions informed by both treatments for traumatic stress and couple approaches to promote trauma healing. Future directions call for increased funding, research, training, and clinical action to treat the effects of traumatic stress on relational family dynamics.     

Stress and subjective life expectancy: Cross-sectional and longitudinal associations in early adolescence

Using data from a longitudinal study, the present study employed a latent class mover-stayer analytical strategy to examine both the cross-sectional and longitudinal (+33 months) relationship between membership of stress classes, and subjective life expectancy. Participants were from 21 High schools in Scotland and Northern Ireland. Data were collected in the first year of High school (mean age = 12.5 years), and at +33 months (fourth year, or school year 11). Sample 1 consisted of 1171 adolescents (40.82% females, 2.56% unreported) in Northern Ireland. Sample 2 consisted of 1059 adolescents (52.79% females, 1.32% unreported) in Scotland. Adolescents with the lowest levels of stress projected the highest subjective life expectancy scores. Longitudinal analyses were jumbled and not in keeping with cross-sectional results. More research may be needed on the trajectory of subjective life expectancy over time before it can be depended upon as a reliable outcome variable in adolescent development.     

Getting older and living up to implicit followership theories: Implications for employee psychological health and job attitudes

Age discrimination at work is a widespread destructive phenomenon that often takes subtle forms. Based on negative stereotypes about older employees, we argue that older employees often experience that they are perceived as less ideal followers than younger employees. We propose that older and younger employees do not differ in what they assume their supervisors expect of an ideal follower (implicit followership theories, IFTs). Thus, we hypothesize that older employees perceive that they compare less favorably to their supervisors' IFTs than younger employees (i.e., worse IFT-fit). This should entail lower quality of the relationship between leader and follower (leader-member-exchange, LMX), which, in turn, should have detrimental effects on employees' health (i.e., burnout) and job attitudes (i.e., job satisfaction, turnover intentions, and identification). We tested our model in a field study with 379 employees. Results differed for ideal and counter-ideal follower attributes. Concerning counter-ideal follower attributes (e.g., being incompetent), age had the proposed negative effect on psychological health and job attitudes that mediated through IFT-fit and LMX. Concerning ideal follower attributes (e.g., thinking ahead), older employees expected their supervisors not only to think less of them than of their younger colleagues, but also expected them to have less demanding IFTs—contrary to our expectations. Employee age was negatively related to psychological health and job attitudes, mediated through lower perceived IFTs, worse perceived appraisal of the actual employee and their joint effects on LMX.