Social and solipsistic observer training: Effects on agreement with a criterion

The need to train accurate, not necessarily agreeing, observers is discussed. Intraobserver consistency as an intermediate criterion in such training is proposed and contrasted with the more familiar criterion of interobserver agreement. Videotaped observations of social interactions between handicapped and nonhandicapped preschoolers provided the medium for examining the criterion agreement of four observers trained against each type of standard. Observers generally failed to show high levels of criterion agreement whether trained to a within- or to a between-observer agreement standard. The results varied somewhat with the frequency of behaviors, however. Correlations between interobserver agreement and intraobserver consistency were variable but somewhat higher when interobserver agreement was the training criterion than when intraobserver consistency was the criterion. Correlations between interobserver agreement and criterion agreement ranged from — .16 to .89 during interobserver agreement training. Correlations between intraobserver consistency and criterion agreement ranged from — .23 to .99 during intraobserver consistency training.     

Briefly presented stimuli can disrupt constant suppression and binocular rivalry suppression

An earlier study has shown that the intermittent suppression of one monocular stimulus by another in binocular rivalry does not occur when normally rivalrous stimuli are briefly presented. Constant suppression of stimuli presented to one eye is a common consequence of esotropia during development. A study is reported which demonstrates that constant suppression does not occur when stimuli are briefly presented. The dependence of suppression on stimulus duration is similar in both forms of suppression.     

Revision of the self-monitoring scale

Snyder's (1974) Self-Monitoring Scale exhibits a stable factor structure that does not correspond to the five-component theoretical structure he presents. Sets of face-valid items that better approximate the theoretical structure are described. Correlations between these sets of items and measures of other constructs reveal that four of the five components are positively related to social anxiety. Effective social interaction is supposedly the high self-monitor's forte, and social anxiety appears to be incompatible with this. The correlational results therefore question the entire theory and indicate the need for a narrower definition of the construct. Adopting such a definition from Snyder's review article (1979), we present a 13-item Revised Self-Monitoring scale which measures only sensitivity to the expressive behavior of others and ability to modify self-presentation. A 20-item Concern for Appropriateness scale is also described. This scale measures 2 variables that are directly associated with social anxiety--cross-situational variability and attention to social comparison information. Both scales have acceptable internal consistency, and both yield 2 subscale scores as well as a total score. Prospective users of either scale are advised to treat the 3 scores separately.     

Punishment of shock-induced aggression

Squirrel monkeys were restrained in a chair equipped with a tail-shock apparatus and a pneumatic bite hose located in front of the subject's face. An aggressive response was recorded when the monkey bit the hose. Initial sessions in which no shocks were delivered produced some biting. When biting during these sessions stabilized at a near-zero level, regularly scheduled shocks were delivered to the monkey's tail, causing a consistently higher rate of biting. After several sessions under these conditions, a punishment phase was introduced in which the previous shock conditions were maintained, and every bite was followed immediately by another, more intense shock. Biting under these conditions was suppressed to a near-zero level. When the punishment contingency was removed, biting increased. With one subject, two additional bite-contingent stimuli were examined: (1) a milder shock that, when made contingent upon hose biting, also suppressed that response, and (2) a contingent tone that had no obvious suppressing or facilitating effect. Individual differences among subjects were extreme, but the effect of bite-contingent shock was consistent. Observations of the subjects during the punishment sessions indicated the existence of certain side effects that resulted from the use of punishment to suppress shock-induced aggression.     

Personality Correlates of Confidence in One's Decisions

ABSTRACT Taylor and Brown (1988) hypothesized that certain illusions, such as unrealistic optimism, are predictably associated with positive affect, social skills, and intellectual functioning. To test this hypothesis, we obtained data from 162 students who first filled out a questionnaire containing dispositional measures of affect, social skills, and approaches to problem solving; in a second session they completed three tasks requiring difficult decisions, reporting their confidence in each decision. Accuracy of judgments was found to vary considerably from task to task, but confidence ratings showed a consistent pattern of individual differences. This result lent support to Taylor and Brown's hypothesis, as did other features of the data, most notably several small but significant Pearson r s between confidence ratings and dispositional variables. Removing the effects of accuracy from these r s reduced their magnitude very little; it yielded partial r s interpretable as evidence that illusory confidence is associated with personality traits which, in this case, load saliently on factors labeled ( a ) affective, and ( b ) cognitive/social.     

Social cues, cognitive style, error magnitude, and male performance on the Felt Figure Replacement Technique.

A series of studies was undertaken to determine possible factors contributing to low predictive utility for Felt Figure Replacement Technique (FFRT) scores from normal, adult males. It was first postulated that normal adult males, compared to normal adult females, are less sensitive to social cues present in the FFRT task. Twenty-six undergraduate males and the same number of undergraduate females were asked to replace both a) department store manikins, thought to be less abstract, and then b) felt figures. While task differences were obtained, these were not dependent on gender, nor did any overall gender differences occur for three types of replacement error. The second study involved 31 college males and 33 college females replacing both felt figures and a male and female confederate. It was assumed that differential attention to social cues would be more visible in this comparison. The general results were identical to the first study.     

Discrimination and generalization by rats of temporal stimuli lasting for minutes

In three experiments, rats learned to respond differentially to reinforced and nonreinforced trials when the length of the intertrial interval (ITI) predicted the trial outcome (reinforcement or nonreinforcement). Rats in control groups, for whom the length of the ITI did not predict the outcome, did not show differential performance on nonreinforced and reinforced trials. Generalization gradients obtained following discrimination training were comparable to those obtained following discrimination training with other types of discriminative stimuli. That is, groups which had shown differential performance in discrimination training yielded generalization gradients with fastest speeds at the previously reinforced ITIs, slowest speeds at the previously nonreinforced ITIs, and intermediate speeds at ITIs of intermediate length. Control groups yielded flat gradients across all ITIs tested. These effects were shown for relative time discrimination (short time = reinforcement, long time = nonreinforcement, or the reverse) and also for an absolute time discrimination (long and short time = reinforcement, middle time = nonreinforcement or the reverse). This method was effective for time duration measured in minutes rather than seconds, as is more commonly the case.