Connectionist Simulation of Attitude Learning: Asymmetries in the Acquisition of Positive and Negative Evaluations

A Connectionist Model of Attitude Formation and Change

This article discusses a recurrent connectionist network, simulating empirical phenomena usually explained by current dual-process approaches of attitudes, thereby focusing on the processing mechanisms that may underlie both central and peripheral routes of persuasion. Major findings in attitude formation and change involving both processing modes are reviewed and modeled from a connectionist perspective. We use an autoassociative network architecture with a linear activation update and the delta learning algorithm for adjusting the connection weights. The network is applied to well-known experiments involving deliberative attitude formation, as well as the use of heuristics of length, consensus, expertise, and mood. All these empirical phenomena are successfully reproduced in the simulations. Moreover, the proposed model is shown to be consistent with algebraic models of attitude formation (Fishbein & Ajzen, 1975). The discussion centers on how the proposed network model may be used to unite and formalize current ideas and hypotheses on the processes underlying attitude acquisition and how it can be deployed to develop novel hypotheses in the attitude domain.

Electrophysiological time course and brain areas of spontaneous and intentional trait inferences

This study measured event-related potentials during spontaneous and intentional trait inferences. Participants read sentences describing the behavior of a target person from which a strong moral trait could be inferred. The last word of each sentence determined the consistency with the trait induced during an introductory paragraph. In comparison with behaviors that were consistent with the implied trait, a P300 waveform was obtained when the behaviors were evaluative inconsistent with that trait. This dependency on behavioral consistency indicates that trait inferences were made previously while reading the preceding behaviors, irrespective of the participants' spontaneous or intentional goals. Overall, the P300 shows considerable parallels between spontaneous and intentional inferences, indicating that the type and timing of the inconsistency process is very similar. In contrast, source localization (LORETA) of the event-related potentials suggest that spontaneous inferences show greater activation in the temporo-parietal junction compared to intentional inferences following an inconsistency. Memory measures taken after the presentation of the stimulus material involved sentence completion and trait-cued recall, and supported the occurrence of trait inferences associated with the actor. They also showed significant correlations with the neural components (i.e. P300 and its current density at the temporo-parietal junction) predominantly following spontaneous instructions, indicating that these components are valid neural indices of spontaneous inferences.

Prejudiced learning: a connectionist account

Connectionist simulation was employed to investigate processes that may underlie the relationships between prior expectancies or prejudices and the acquisition of attitudes, under conditions where learners can only discover the valence of attitude objects through directly experiencing them. We compared contexts analogous to learners holding either false negative expectancies ('prejudices') about a subclass of objects that were actually good or false positive expectancies about objects that were actually bad. We introduced expectancy-related bias either by altering the probability of approach, or by varying the rate of learning following experience with good or bad objects. Where feedback was contingent on approach, the false positive expectancies were corrected by experience, but negative prejudices resisted change, since the network avoided objects deemed to be bad, and so received less corrective feedback. These findings are discussed in relation to the effects of intergroup contact and expectancy-confirmation processes in reducing or sustaining prejudice.

Attitudes as Object-Evaluation Associations of Varying Strength

Historical developments regarding the attitude concept are reviewed, and set the stage for consideration of a theoretical perspective that views attitude, not as a hypothetical construct, but as evaluative knowledge. A model of attitudes as object-evaluation associations of varying strength is summarized, along with research supporting the model's contention that at least some attitudes are represented in memory and activated automatically upon the individual's encountering the attitude object. The implications of the theoretical perspective for a number of recent discussions related to the attitude concept are elaborated. Among these issues are the notion of attitudes as "constructions," the presumed malleability of automatically-activated attitudes, correspondence between implicit and explicit measures of attitude, and postulated dual or multiple attitudes.

Valence, Comparison Focus and Self-Positivity Biases

Abstract. The present research questions whether mere valence affects self-other comparisons in the domain of trait characteristics. While some previous studies have reported greater positivity bias for the self when traits were positive than when traits were negative, we suggest that this is an ambiguous finding, because valence and content were confounded. When we unconfounded content and valence, valence had no effect on the magnitude of self-positivity bias displayed. We also replicate several findings for our unconfounded set of traits. Firstly, comparing others to the self, rather than comparing the self to others, lowered self-positivity for positive and negative traits (focus effect). Secondly, extremely positive and negative traits triggered greater positivity bias than did more moderately evaluated ones. Finally, we suggest that comparative self-positivity biases may be based on a general positivity bias.

Why Most People Disapprove of Me: Experience Sampling in Impression Formation

Individuals are typically more likely to continue to interact with people if they have a positive impression of them. This article shows how this sequential sampling feature of impression formation can explain several biases in impression formation. The underlying mechanism is the sample bias generated when the probability of interaction depends on current impressions. Because negative experiences decrease the probability of interaction, negative initial impressions are more stable than positive impressions. Negative initial impressions, however, are more likely to change for individuals who are frequently exposed to others. As a result, systematic differences in interaction patterns, due to social similarity or proximity, will produce systematic differences in impressions. This mechanism suggests an alternative explanation of several regularities in impression formation, including a negativity bias in impressions of outgroup members, systematic differences in performance evaluations, and more positive evaluations of proximate others.

Attitude formation through exploration: valence asymmetries

The formation of attitudes toward novel objects was examined as a function of exploratory behavior. An initial experiment, in which participants played a computer game that required them to learn which stimuli, when sampled, produced favorable or unfavorable outcomes, demonstrated learning, attitude formation, and generalization to novel objects. The findings also revealed 2 interesting valence asymmetries: a learning asymmetry involving better learning for negatively valenced than positively valenced objects and a generalization asymmetry involving stronger generalization as a function of negative than of positive attitudes. Findings from 4 experiments led to an explanation of the learning asymmetry in terms of information gain being contingent on approach behavior and related the generalization asymmetry to a negativity bias that weighs resemblance to a known negative more heavily than resemblance to a positive.