Labels as features (not names) for infant categorization: a neurocomputational approach

A substantial body of experimental evidence has demonstrated that labels have an impact on infant categorization processes. Yet little is known regarding the nature of the mechanisms by which this effect is achieved. We distinguish between two competing accounts: supervised name-based categorization and unsupervised feature-based categorization. We describe a neurocomputational model of infant visual categorization, based on self-organizing maps, that implements the unsupervised feature-based approach. The model successfully reproduces experiments demonstrating the impact of labeling on infant visual categorization reported in Plunkett, Hu, and Cohen (2008) . It mimics infant behavior in both the familiarization and testing phases of the procedure, using a training regime that involves only single presentations of each stimulus and using just 24 participant networks per experiment. The model predicts that the observed behavior in infants is due to a transient form of learning that might lead to the emergence of hierarchically organized categorical structure and that the impact of labels on categorization is influenced by the perceived similarity and the sequence in which the objects are presented. The results suggest that early in development, say before 12 months old, labels need not act as invitations to form categories nor highlight the commonalities between objects, but they may play a more mundane but nevertheless powerful role as additional features that are processed in the same fashion as other features that characterize objects and object categories.     

Effects of positive and negative feedback on behavior control in hyperactive and normal boys

The hypothesis that hyperactive boys have relatively less response to negative feedback than to positive feedback was studied. Sixteen hyperactive boys and 16 controls were compared on two tasks under different feedback conditions. Feedback conditions were no feedback, positive feedback, and negative feedback. Tasks were symbol encoding and correcting spelling words. Hyperactives and controls were compared in amount of time on-task and amount of work correctly completed. Hyperactives were on-task significantly more under conditions of negative feedback than under positive feedback, but negative feedback significantly increased errors on the spelling correction task. Controls were equally responsive to positive, negative, or no feedback. Hyperactives accomplished significantly less than controls on the coding task, but performed as well as controls on the spelling correction task, which was administered to each boy at his own level of spelling ability. The results imply that while consistent negative feedback can reduce off-task behavior for hyperactives, it can also decrease the accuracy of the work they are doing.This research was supported in part by U.S. Public Health Service Training Grant (in Biological Science) No. MH07081. This article is based on a dissertation presented to the Graduate School of Arts and Sciences of Washington University in partial fulfillment of the requirements for the Ph. D. degree. The assistance of Dr. John A. Stern in helping with the preparation of the dissertation is gratefully acknowledged. The generous assistance of Dr. Cynthia Janes in helping prepare this paper is appreciated. The dissertation is available from University Micro-films (Order No. 74-13, 799). The assistance of Ms. B. Talent and Ms. S. Weiner in making reliability checks is gratefully acknowledged.     

Performance and activity of hyperactive and normal boys as a function of distraction and reward

Twenty-five hyperactive boys and 25 controls matched for age, social class, and race were compared on three performance tasks (coding, tone discrimination, and connecting dots) in two settings (nondistracting and highly distracting). Control Ss performed significantly better than did hyperactives in all conditions, except during tone discrimination and connecting dots in the nondistracting setting. Distraction decreased performance for both groups on the coding task and for hyperactives on the tone discrimination task, but significantly improved performance for controls on the connecting dots task. Distraction, where detrimental, was not significantly more so for hyperactives than for controls. The effect of reward on coding performance was studied in the distracting condition; reward produced the best performance for both groups, but significant differences between groups were not found.This research was supported in part by U.S. Public Health Service Training Grant (in Biological Sciences) No. MH07081.The participation of the following people in helping to design this research study and collect data is hereby acknowledged: David Bremer, Ted Goltz, Ruth Rosenthal, and Michael Stern.     

Catching up with wonderful women: The women‐are‐wonderful effect is smaller in more gender egalitarian societies

Inequalities between men and women are common and well‐documented. Objective indexes show that men are better positioned than women in societal hierarchies—there is no single country in the world without a gender gap. In contrast, researchers have found that the women‐are‐wonderful effect—that women are evaluated more positively than men overall—is also common. Cross‐cultural studies on gender equality reveal that the more gender egalitarian the society is, the less prevalent explicit gender stereotypes are. Yet, because self‐reported gender stereotypes may differ from implicit attitudes towards each gender, we reanalysed data collected across 44 cultures, and (a) confirmed that societal gender egalitarianism reduces the women‐are‐wonderful effect when it is measured more implicitly (i.e. rating the personality of men and women presented in images) and (b) documented that the social perception of men benefits more from gender egalitarianism than that of women.     

Coping with dialects from birth: Role of variability on infants’ early language development. Insights from Norwegian dialects

Previous research suggests that exposure to accent variability can affect toddlers’ familiar word recognition and word comprehension. The current preregistered study addressed the gap in knowledge on early language development in infants exposed to two dialects from birth and assessed the role of dialect similarity in infants’ word recognition and comprehension. A 12-month-old Norwegian-learning infants, exposed to native Norwegian parents speaking the same or two Norwegian dialects, took part in two eye-tracking tasks, assessing familiar word form recognition and word comprehension. Their parents’ speech was assessed for similarity by native Norwegian speakers. First, in contrast to previous research, our results revealed no listening preference for words over nonwords in both monodialectal and bidialectal infants, suggesting potential language-specific differences in the onset of word recognition. Second, the results showed evidence for word comprehension in monodialectal infants, but not in bidialectal infants, suggesting that exposure to dialectal variability impacts early word acquisition. Third, perceptual similarity between parental dialects tendentially facilitated bidialectal infants’ word recognition and comprehension. Forth, the results revealed a strong correlation between the raters and parents’ assessment of similarity between dialects, indicating that parental estimations can be reliably used to assess infants’ speech variability at home. Finally, our results revealed a strong relationship between word recognition and comprehension in monodialectal infants and the absence of such a relationship in bidialectal infants, suggesting that either these two skills do not necessarily align in infants exposed to more variable input, or that the alignment might occur at a later stage.