Edmund S. Howe’s research while affiliated with University of Missouri–St. Louis and other places

This experiment explored the dimensionality of judgments of visual patterns varying in form and amount of whole and part symmetry. The study compared judgments made by independent subjects on scales of symmetry, balance, pleasingness, goodness, simplicity, organization, and dispersion. All judgments showed very high reliabilities. A principle factors analysis of the interscale correlation matrix yielded a substantial single factor of symmetry/pleasingness upon which all scales but one (dispersion) were very highly loaded. The structure of the main factor indicates that, regardless of the nominal scale instructions to raters, the results are very closely equivalent. Dispersion judgments were highly reliable but were uncorrelated with the main factor.

The superior reproducibility of good, symmetrical patterns (the pattern goodness effect) is well documented. In the present experiment, the temporal course of the phenomenon was explored. Subjects attempted to recall 64 dot patterns having Array Sizes 4−7 and four levels of symmetry, following backward-masked exposure. It was found that all patterns were processed at the same rate, as is shown by a constant slope for all array size × symmetry × exposure time functions. An empirical index of pattern nameability was quite strongly correlated with symmetry and goodness. For Array Size 7 patterns, inclusion of the nameability index significantly raised the predictability of pattern recall above that observed with goodness value alone. These results suggest that when the immediate memory span is exceeded, the visual code may be supplemented by a verbal one.

This article reports two experiments requiring subjective evaluative judgments of the potential dangerousness of hypothetical persons. The research operationally fits the paradigm for the study of personality impression formation, and seeks to illuminate the processes by which two offenses combine to evoke a net judgment of dangerousness. The theoretical framework and philosophy adopted is Anderson's information integration and functional measurement theory. In Study 1, all paired combinations of 10 distinctive crimes were each presented as having been committed by the same person on two separate occasions. Subjects judged overall offender dangerousness. In Study 2, judgments of dangerousness were made when the time purportedly elapsing between two crimes was systematically varied over several ranges of up to 41 years. Three key findings emerged. First, judgments of dangerousness result from an averaging process. This result yields paradoxical implications having considerable pragmatic significance. Second, judgments of dangerousness following two sequential criminal acts (one of high and one of low seriousness) are consistently higher when the high seriousness one is the second crime. Third, with certain qualifications discussed in the text, a serious earlier crime appears to elicit an approximately constant magnitude of judged present dangerousness no matter how long ago it was perpetrated. This result implies that subjects infer considerable permanence of criminal predilection to those who have committed a serious crime in the past.

The purposes of these studies were (a) to establish on a general basis whether 3 time-honored manipulable components of punishment combine additively, or multi-plicatively in determining the judged deterrence value of a sanction, and (b) to explore the equivalence in principle of two mutually exclusive experimental designs. Study 1 used a repeated measures design whereas main and interactive effects of certainty and severity were substantial and appeared to follow a multiplying rule, celerity effects were very small and local. Studies 2a and 3 used independent groups and established that the results of Study 1 were not merely an arbitrary product of the method. The discussion addresses the minimal effects of celerity, the interaction between certainty and severity, and the theoretical implications of choice of method.

In two experiments the subjects, who included State Circuit Court Judges, assigned a quantitative judgment of blame to hypothetical offenders as a function of level of intention and seriousness of outcome. In accordance with the findings of numerous other investigators using the Intent x Damage paradigm, averaged subjects appeared to follow an adding rule for the integration of intention and outcome damage information. Furthermore, and in keeping with a key methodological purpose of the research, equivalent averaged results were observed whether student subjects made just one judgment of a single Level of Intention x Level of Outcome Damage treatment combination (independent groups design) or multiple judgments of all possible such factorial treatment combinations (repeated measures design). However, statistical analysis of individual students' and individual judges' repeated measures showed that about half of them adopted an adding rule whereas the judgments of others conformed to an intention-only rule. Some implications of the findings are: (a) in critical respects students' results are quite generalizable since there were no striking student versus judge differences; (b) averaged outcomes conceal important interpretable individual differences in information processing that are highly commensurate for students and judges; and (c) there is thus a defensible and pragmatically desirable economy in the adoption of repeated measures designs for the study of individual persons in such less accessible categories as the Judiciary.

This article examines the problem of how undergraduate students and State Circuit Court Judges combine three specific components of information (pertaining to mitigation, intention, and seriousness of outcome damage) to make a net evaluative judgment of blame. The problem is considered within the framework and philosophy of Anderson's information integration and functional measurement theory. In three experiments, two levels each of the three information components identified were factorially crossed for two independent cases: self-defense under provocation, and altruistic defense of another. Target stimuli consisted of 100-word scenarios each embodying one factorial treatment combination. Experiment 1 used an independent group and Experiment 2 a repeated measures design. Experiment 3 replicated the second experiment using Judges as subjects. It was found that students integrated mitigation, intention, and outcome damage information by an adding process, and that the results concerning these three independent variables were essentially similar for both types of design. The most striking differences between Judges' and students' judgments centered on the differential use of mitigation information. Judges showed a far broader range of nonadditive, simple interaction effects between mitigation and both outcome damage and type of case. These results presumably reflect the primary concerns and responsibilities of the Judiciary. Questions of mitigation may thus complicate the otherwise straightforward additive integration of information in the intent-damage paradigm, and in judgments of seriousness.

Numerous studies have concluded that the judged probability of conviction for a crime is higher when information concerning a prior conviction is disclosed. This is even more true when the prior and present crimes are the same. This paper examines the argument that the influence of a hypothetical previous conviction on the judgment of predilection toward a hypothetical present crime is a continuous function of the degree of similarity between past and present crimes. The subjects were presented with all possible pairs of 10 crimes, the first being considered as a prior conviction and the second as a presently charged crime. The subjects judged the subjective likelihood that a person convicted of the first crime (C1) would, in fact, he of a mind and inclination to commit the second one (C2). Independent judgments of intercrime, undefined similarity also were obtained. The main experimental findings, as expected, were that the judged probability of C2. given C1, was: (a) greatest when C2 was a repeat of C1; (b) uniquely different for each C1; (c) predicted quite well by degree of intercrime similarity; (d) poorly predicted by crime seriousness values. Thus, the subject appears to make judgments of predilection on the basis of simple representativeness heuristics, which specify that certain crimes will be considered more likely if they are more semantically related to earlier ones. Such a mental device unfortunately would he systematically biased in courtroom settings because judgments of intercrime similarity are not influenced by real-world probabilities.

The certainty and severity of punishment for crimes are commonly given credence as determinants of deterrence value, whereas celerity of punishment is not. Moreover, data are sparse and divided on the question of how certainty and severity components combine algebraically. Two experiments inspected the effects of certainty, severity and celerity of hypothetical punishments on judged deterrence value, and the form of their factorial combination. Judged deterrence scale values were obtained for eight hypothetical conditions of punishment for serious crimes. These conditions of punishment consisted of orthogonal combinations of two levels each of certainty, severity, and celerity and were administered to independent subjects. Strong effects of certainty and severity and moderate effects of celerity were found, and there were no interactions among the three variables. Thus, celerity is pertinent to judged deterrence value, and the three components of punishment clearly combine additively rather than multiplicatively. It is argued that despite the empirical results, certainty, severity, and celerity must, however, ultimately be showp to combine according to a multiplying rule. The Discussion centers largely on an analysis and justification for that argument.