ON LEARNING TO PREDICT EVENTS

B jörkman , M. On learning to predict events. Scand. J. Psychol ., 1962, 3 , 233–240.—A rather large group of 'real life situations' requires a subject to predict events which are causally related to variables of the situation. Some operational characteristics of such situations are described. A primary question concerns the learning of the connection between situation and event ( S → E learning).—In an illustrative experiment frequency of 'reinforcement by observation' is varied. It is found that this factor can vary considerably without affecting rate of learning. As a preliminary conclusion it is suggested that reinforcement by observation as a differentiating factor is dependent on other factors, e.g. the dispersion of states of the situation.     

THE TIME-ORDER ERROR IN THE CONSTRUCTION OF A SUBJECTIVE TIME SCALE

(1) A method to obtain a scale of subjective time unbiased by the time-order error is suggested. (2) The method gives as by-product some preliminary results on the memory process which gives rise to the time-order error. (3) The method is applied in an experimental study. Two methods, the adjustment method and the reproduction method were used to construct a subjective time scale within the range 1–7 seconds. The former method gave an almost linear R-S curve while the latter resulted in a positively accelerated curve. (4) The time-order error seems to be a very rapid process. The memory curve reaches very near its asymptotic value within a few tenths of a second.     

STUDIES IN PREDICTIVE BEHAVIOR: Explorations into predictive judgments based on functional learning and defined by estimation, categorization, and choice

A model for functional learning with a subjective regression which intervenes between stimuli and predictive responses was used. The subject had to predict motion time of a ball rolling down a chute. The estimation experiments showed that (1) the relation between predicted and objective time, is biased in the direction of t _pred= t ², (2) training does not increase the veridicality. In the categorization experiments (1) learning occurred more rapidly with homogeneous categories, (2) transfer was facilitated after training on heterogeneous categories, (3) learning occurred without reinforcement by observation. In a choice experiment two simultaneous processes were studied, functional and probability learning.     

Apparent impropriety and procedural fairness judgments

Variation in decision making and allocation procedures has been shown to affect judgments of the fairness of the procedure and its outcome, but such effects have always been studied in the context of properly enacted procedures. It was hypothesized that the appearance of impropriety in the enctment of a fair procedure would increase the extent to which the procedure is judged in terms of its outcome. One hundred twenty undergraduate males and females were placed in the role of either defendant or observer with respect to an adversary procedure trial. Appearance of impropriety was manipulated during the trial by either including or not including evidence of a friendly personal relationship between the judge and the plaintiff's lawyer. The defendant was said to have either won or lost the case. A significant impropriety × outcome interaction on ratings of procedural fairness, unqualified by higher order effect, supported the hypothesis: a favorable outcome increased and an unfavorable outcome decreased the fairness of the procedure more when the impropriety was present. Discussion focuses on the implications of these findings for future investigation and theory on procedural justice and for practical issues.     

Where should we draw the line between quality of care and other ethical concerns related to medical registries and biobanks?

Together with large biobanks of human samples, medical registries with aggregated data from many clinical centers are vital parts of an infrastructure for maintaining high standards of quality with regard to medical diagnosis and treatment. The rapid development in personalized medicine and pharmaco-genomics only underscores the future need for these infrastructures. However, registries and biobanks have been criticized as constituting great risks to individual privacy. In this article, I suggest that quality with regard to diagnosis and treatment is an inherent, morally normative requirement of health care, and argue that quality concerns in this sense may be balanced with privacy concerns.     

Beware how you compare: comparison direction dictates stimulus-valence-modulated presentation-order effects in preference judgment

Englund and Hellström (Journal of Behavioral Decision Making 25:82–94, 2012a) found a tendency to prefer the left (first-read) of two attractive alternatives but the right (second-read) of two unattractive alternatives—a valence-dependent word-order effect (WOE). They used stimulus pairs spaced horizontally, and preference was indicated by choosing one of several written statements (e.g., “apple I like more than pear”). The results were interpreted as being due to stimulus position, with the magnitude of the left stimulus having a greater impact on the comparison outcome than the magnitude of the right. Here we investigated the effects of the positioning of the stimuli versus the semantics of the response alternatives (i.e., comparison direction) on the relative impacts of the stimuli. Participants rated preferences for stimuli spaced horizontally with the response alternatives not dictating a comparison direction (Exp. 1), and stimuli spaced vertically using Englund and Hellström’s (Journal of Behavioral Decision Making 25:82–94, 2012a) response alternatives, which dictate a comparison direction semantically (Exp. 2). The results showed that the valence-dependent WOE found by Englund and Hellström (Journal of Behavioral Decision Making 25:82–94, 2012a) was not due to the horizontal stimulus positioning per se, but to the induced comparison direction, with the effect probably being mediated by attention directed at the subject of the comparison. We concluded that a set comparison direction is required for the valence-dependent WOE to appear, and that using Hellström’s sensation-weighting model to determine stimulus weights is a way to verify the comparison direction.     

Predictors of outcome in Internet-based cognitive behavior therapy for severe health anxiety

Internet-based cognitive behavior therapy (CBT) for severe health anxiety can be effective, but not all patients achieve full remission. Under these circumstances, knowledge about predictors is essential for the clinician in order to make reliable treatment recommendations. The primary aim of this study was to investigate clinical, demographic, and therapy process-related predictors of Internet-based CBT for severe health anxiety. We performed three types of analyses on data from a sample comprising participants (N = 81) who had received Internet-based CBT in a randomized controlled trial. Outcomes were a) end state health anxiety, b) improvement in health anxiety (continuous change scores), and c) clinically significant improvement. Outcomes were assessed at six-month follow-up. The results showed that the most stable predictors of both end state health anxiety and improvement were baseline health anxiety and depressive symptoms. Treatment adherence, i.e. the number of completed treatment modules, also significantly predicted outcome. Notably, health anxiety at baseline was positively associated with symptom improvement while depressive symptoms was negatively related to improvement. Demographic factors were largely without significant impact on end state symptoms or improvement. We conclude that baseline symptom burden and adherence to treatment have strong predictive effects in Internet-based CBT for severe health anxiety.     

Poor motor skills: A risk marker for bully victimization

Children who are clumsy are often bullied. Nevertheless, motor skills have been overlooked in research on bullying victimization. A total of 2,730 Swedish adults (83% females) responded to retrospective questions on bullying, their talents in physical education (i.e., coordination and balls skills) and school academics. Poor talents were used as indicators of poor gross motor skills and poor academic skills. A subset of participants also provided information on educational level in adulthood, childhood obesity, belonging to an ethic minority in school and socioeconomic status relative to schoolmates. A total of 29.4% of adults reported being bullied in school, and 18.4% reported having below average gross motor skills. Of those with below average motor skills, 48.6% were bullied in school. Below average motor skills in childhood were associated with an increased risk (OR 3.01 [95% CI: 1.97–4.60]) of being bullied, even after adjusting for the influence of lower socioeconomic status, poor academic performance, being overweight, and being a bully. Higher odds for bully victimization were also associated with lower socioeconomic status (OR 2.29 [95% CI: 1.45–3.63]), being overweight (OR 1.71 [95% CI: 1.18–2.47]) and being a bully (OR 2.18 [95% CI: 1.53–3.11]). The findings indicate that poor gross motor skills constitute a robust risk‐marker for vulnerability for bully victimization. Aggr. Behav. 39:453–461, 2013. © 2013 The Authors. Aggressive Behavior Published by Wiley‐Blackwell