Evidence for episodic memory in a Pavlovian conditioning procedure in rats

O'Brien, Jamus; Sutherland, Robert J. Evidence for episodic memory in a Pavlovian conditioning procedure in rats. Hippocampus. Vol 17(12) 2007, 1149-1152.

Several requirements have been proposed for establishing episodic memory in nonlinguistic species. Clayton et al. (2003) suggest that episodic memory competence requires integrated representations of “what", "where", and "when” content that can be flexibly updated as more information is gathered. Tests of episodic memory should be novel and unexpected, exceed the capacity of short-term memory (Dere et al., 2006), be unsolvable by familiarity judgments (Gallistel, 1990), and involve memories formed in unique one-trial learning episodes (Morris, 2001). In this article, O’Brien and Sutherland design and run an experiment that attempts to meet these requirements while testing for the flexible and integrated representations required for episodic memory.

In this experiment, twenty-two male Long-Evans rats were run through three experimental phases. In Phase 1, rats were given time to explore two different “contexts” (boxes A and B). These boxes were Plexiglas modular test chambers with steel grid floors. Box A had black colored walls and was scented with Quatzyl-D-Plus, while box B was white and scented with Clinicide. Each rat visited Box A on three successive mornings and Box B on three successive evenings. In Phase 2, rats were exposed to a single immediate shock in a “chimerical” box that was half black, half white, and unscented. For half the rats, this occurred in the morning and for the other half, it occurred in the evening; this is the study’s independent variable. In Phase 3, during mid-day, rats were placed in one of the contexts from Phase 1 (either Box A or Box B). Fear responses (this study’s dependent variable) were then measured by timing conditioned “freezing” behavior.

A repeated measure ANOVA was conducted on the percent of time spent freezing during Phase 3, and it was predicted that rats placed in the context congruent to the time of day of their shock would exhibit more freezing than rats placed in the incongruent context. (For example, if a rat was in Box A in the morning and Box B in the evening during Phase 1 and received its shock in the morning during Phase 2, then the congruent context would be Box A and the incongruent context would be Box B.) The results were a significant relationship in the direction predicted (F_(1,20) = 45.0, P < 0.001).

These findings support the idea that rats acquire memories laden with temporal context (or “when” content), such as the time of day, which is an important requirement of establishing episodic memory competence. Further studies could continue this line of research by establishing evidence for rats generating “what” and “where” content, and by exploring the nature of rats’ temporal cues – possibly endogenous circadian oscillators (Gallistel, 1990) or the age of granule cells in the dentate gyrus of the hippocampus (Aimone et al., 2006).

Web-based Methods in Terrorism and Disaster Research

Schlenger, W.E. & Silver, R.C. (April 2006). Web-based Methods in Terrorism and Disaster Research. Journal of Traumatic Stress, Vol. 26, No. 2, 185-193.

Gaining access to a traumatized population in the aftermath of a disaster can be challenging. The need for rapid response and appropriate probability sampling along with the observational nature of the studies (i.e. no random assignment) and post-only design can be problematic for the generalizability of the results. Recently, web-based methods have helped to address some of these age-old issues. (1) The literature indicates that people respond more honestly to sensitive questions in self-report than in interview-based assessments. (2) The use of e-mail also assists greatly with retention rates in longitudinal studies. (3) Respondents can answer surveys within the privacy of their home at a time that is convenient to them. (4) Question delivery can be standardized. (5) Time-consuming and error-prone steps of data coding and entry are eliminated. (6) And most importantly, recent companies such as Knowledge Networks Inc. have recruited nationally representative probability samples for just such uses. Panels can even be created in advance of disasters, enabling premeasure to be linked to postevent responses. However, these new technologies are not without their issues, e.g. populations may suffer infrastructure disruptions due to disaster that render data collection impossible such as in Hurricane Katrina.