Environmental context-dependent eyewitness recognition

The present study helped resolve the apparent conflict between many laboratory list-learning studies, which have not found environmental context-dependent recognition memory, and staged field studies (e.g. Malpass and Devine, 1981), whose results with ‘guided memory’ techniques suggest that eyewitness face recognition should depend upon environmental context reinstatement. It was found in two different experiments that, relative to testing in a new place, returning participants to the environment where a live staged event had occurred improved performance on identification of a confederate's face (i.e., hit rate). Although physical reinstatement improved identification performance in Experiment 1, mental reinstatement instructions to subjects tested in a new environment did not improve identification performance over an uninstructed group. The environmental reinstatement effect did not interact with test delay or confederate. In Experiment 2 it was found that environmental reinstatement improved accuracy (hit rate and foil identification rate) when the correct target was present in the test line-up, and that false identifications were not significantly affected by contextual manipulations when the correct target was absent from the line-up. The results provide an empirical basis for the hypothesis that returning to the scene of an event improves eyewitness face recognition.     

Parallel distributed processing and neuropsychology: A neural network model of Wisconsin card sorting and verbal fluency

Neural networks can be used as a tool in the explanation of neuropsychological data. Using the Hebbian Learning Rule and other such principles as competition and modifiable interlevel feedback, researchers have successfully modeled a widely used neuropsychological test, the Wisconsin Card Sorting Test. One of these models is reviewed here and extended to a qualitative analysis of how verbal fluency might be modeled, which demonstrates the importance of accounting for the attentional components of both tests. Difficulties remain in programming sequential cognitive processes within a parallel distributed processing (PDP) framework and integrating exceedingly complex neuropsychological tests such as Proverbs. PDP neural network methodology offers neuropsychologists co-validation procedures within narrowly defined areas of reliability and validity.     

Psychology ethics down under: a survey of student subject pools in Australia

A survey of the 37 psychology departments offering courses accredited by the Australian Psychological Society yielded a 92% response rate. Sixty-eight percent of departments employed students as research subjects, with larger departments being more likely to do so. Most of these departments drew their student subject pools from introductory courses. Student research participation was strictly voluntary in 57% of these departments, whereas 43% of the departments have failed to comply with normally accepted ethical standards. It is of great concern that institutional ethics committees apparently continue to condone, or fail to act against, unethical research practices. Although these committees have a duty of care to all subjects, the final responsibility for conducting research in an ethical manner lies with the individual researcher.     

Brief report: An extension of the sequential alternating treatment design using reversals with subjects not available concurrently

This brief report proposes an extension of the traditional Alternating Treatments Design that would be acceptable to use in hospital and residential settings where subjects are often not concurrently available. Concerns about the inability of traditional designs to work in these special situations and the emergence of a “scientist–practitioner split” are discussed. An example of how the Extended Alternating Treatments Design might be used is provided.     

Convergent methods in psychomusical research based on integrated,interactive computer control

Much of what we learn from an experiment depends on what we ask of the subjects and how we ask it. Most music perception and cognition research is implemented using a single methodology. Recently, questions have been raised about the validity and generalizability of such a limited approach, particularly regarding such complicated theoretical constructs as musical tonality. We propose that a useful technique is to converge on the answer to experimental questions by applying multiple methods, in essence, simultaneously investigating the central research question as well as ancillary questions of method. Such an approach is facilitated by a computer hardware and software system that integrates experimental design, data collection, data analysis, and data dissemination, linking off-the-shelf components through standard file formats and data-exchange protocols. The list-based system, using an IBM-type PC running Windows 3.0, handles real-time natural sound sampling and playback, signal synthesis and analysis, MIDI input and output, digitized picture display, and various subject-response mechanisms. On the basis of graphical icons, the integrated system can be used to build custom experimental designs. Examples of convergent psychomusical experiments that were designed, implemented, and analyzed using the system are discussed.     

The validity of verbal protocols

The reactivity of a "think aloud" verbal protocol and the veridicality of different retrospective protocols were tested over four dissimilar tasks. Generating a concurrent protocol altered accuracy in two tasks, simple addition and a choice between two gambles, and generally prolonged response times. Such reactivity partially qualifies the dominant theory of protocol generation (Ericsson & Simon, 1984). Retrospective protocols yielded substantial forgetting or fabrication in all tasks, supporting the consensus on the nonveridicality of these methods. It is concluded that protocol validity should be based on an empirical check rather than on theory-based assurances.     

Employment demographics and attitudes that predict preferences for alternative pay increase policies

A survey of 942 university staff employees shows that performance and cost of living are preferred over seniority as criteria upon which to base pay increases. Multiple regression analysis reveals that preferences for performance-based pay increases are most strongly related to perceptions that merit pay does not decrease co-worker cooperation, perceptions that performance differences exist among co-workers, employment in white collar jobs, and never having belonged to a union. Preferences for seniority-based and cost of living-based pay increases are most strongly related to the perceived need for a union, perceptions that merit pay reduces cooperation, education level, and gender. In addition, preferences for seniority-based pay increases are related to years of service.An earlier version of this paper was presented at the 1989 meetings of the Midwest Academy of Management.