The Editors encourage submission of any announcements, and/or letters to the editors, regarding psychological science.
Comments on the content and presentation of the newsletter are also appreciated.
Kristi S. Multhaup
Mark E. Faust
Univ. of North Carolina at Charlotte
Why waste your time subjecting your family and friends to your humor when you can elicit guffaws from your colleagues? Send us your science related humor: email@example.com
Subscribe to the Division 3 E-mail network to keep informed about Division 3 and issues regarding psychological science. This is a monitored network to keep the number of e-mails down.
Subscribe: Send an e-mail to firstname.lastname@example.org. Leave the Subject line blank and type “subscribe div3” in the body of the message.
Send a Message (once subscribed): email@example.com
Questions: Send e-mail to Neal Johnson, Ohio State University, firstname.lastname@example.org
Thomas R. Zentall
University of Kentucky
Johns Hopkins University
University of Colorado
Wilfrid Laurier University
Charles L. Brewer
Members-At-Large of the
Mark A. McDaniel (8/05-08)
Washington University, St. Louis
Valerie F. Renya (8/05-08)
Nelson Cowan (8/04-07)
University of Missouri
Ralph R. Miller (8/04-07)
Binghamton Univ., SUNY
Nelson Cowan (8/04-07)
University of Missouri
Mark H. Ashcraft (8/05-06)
Veronica J. Dark (8/03-06)
Iowa State University
Representative to APA Council
Lewis P. Lipsitt (8/04-07)
Emanuel E. Donchin (8/03-06)
University of South Florida
Board of Directors
J. Bruce Overmier
University of Minnesota
William D. Timberlake (Awards)
Mark H. Ashcraft (Fellows)
Randall W. Engle (Membership)
Georgia Institute of Technology
Marvin Lamb (Program)
Cal. State Hayward
One of the important differences between research with humans and with other animals is humans generally come to the experiment with elaborative ‘pretraining’ involving experience with many tasks and with an excellent command of language. The advantage of task experience is that these subjects are prepared for a challenging task and they have already acquired most of the skills necessary to acquire or perform it. The advantage of having an excellent command of a language is the experimenter can tell them what they are supposed to do (i.e., give them instructions). Many
researchers working with humans do not sufficiently appreciate the role of instructions in structuring the tasks they give subjects and determining the range of behavior that they will evoke. In some cases instructions are critical to avoid acquisition failure. In other cases they may avoid long periods of training while the subjects are learning what the experimenter has in mind.
With animals there is no choice. The task instructions must be incorporated into the task and acquired by trial and error. Often the problem with the inability to procedurally separate the failure to acquire a task from the failure of task instructions is that when animals perform poorly it is generally concluded that they do not have the necessary cognitive ability necessary to master the task rather than that the instructions were ambiguous.
An example may clarify this point. Some time ago, learning researchers were interested in the effects of list interference on the memory for lists of words. Subjects acquired a list of words and were then given a second list of words to remember. They could then be asked to recall items either from the first or the second list, either immediately or after some time had passed. The typical finding was that on an immediate test memory for items from the second list was very good but declined with the passage of time, however, initially memory for items from the first list was poor but it improved with time. The argument for this recovery was that initially items from the second list interfered with memory for items from the first list, but with the passage of time, memory for items from the first list spontaneously recovered.
Interestingly, research with rats produced quite similar results and thus, appeared to support the spontaneous recovery interpretation; however, it can be argued that there is an important procedural difference between the research with humans and other animals. At the time of recall the human subjects were given explicit instructions to recall one list or the other, whereas the rats were not given instructions. In fact, for the rats, typically the task one and task two responses were mutually exclusive (e.g., one was a right turn and the other a left turn in a T-maze). Thus, the changes in performance as a function of the retention interval may not have reflected the loss of memory at all but the ambiguity that came from the absence of instructions – specifically, the high probability of a task two response on an immediate test, but the reemergence of a task one response with increasing retention interval. That is, on an immediate test the last response made is likely to be correct whereas on a delayed test, it is not clear which response is more likely to be correct.
To test this hypothesis, I repeated the rat experiment but provided differential distinctive contextual cues during the acquisition of the two responses (Zentall, 1970). I could then reintroduce either set of those contextual cues on test trials to signal to the animals which response they were supposed to make. On an immediate test, nondifferentially cued control rats showed the typical bias to make the second response acquired and this bias dissipated as the retention interval increased. However, experimental rats that were cued to recall the response from one task or the other showed no bias and no forgetting, even after a 30 day retention interval!
More recently, we have applied this notion of instructional ambiguity the assessment of memory in matching-to-sample tasks with pigeons. With this task, pigeons are originally trained to associate one initial stimulus or sample with reinforced responding to one comparison stimulus and to associate a different sample with reinforced responding to the other comparison stimulus. Memory for the sample can then be assessed by inserting a delay (that typically varies in duration from trial to trial) between the sample and the comparison stimuli. In most matching-to-sample research, the appearance of the delay is very similar to the appearance of the intertrial interval (i.e., all lights off). If the pigeon ‘misinterprets’ the delay as an intertrial interval, the decline in matching performance with increasing delay may be difficult to interpret. Could the test trial have been a trial without a sample? The problem is exacerbated if the delay actually resembles a sample event. For example, in the case of present/absent sample matching, on half of the trials a sample is presented and one comparison stimulus is correct, whereas on the remaining trials there is no sample and the other comparison is correct. If one now introduces a delay between the sample and the comparisons, one might expect pigeons to show a bias to choose the comparison associated with the absent sample because now on all trials no sample appears immediately preceding the comparison stimuli. And that is exactly what has been found. One way to avoid this ambiguity problem is to make all of the events that might occur on a trial dissimilar from all of the other trial events (Sherburne Kaiser, & Zentall, 1998). Such a procedure does in fact eliminate the ‘absent’ bias that has typically been found. But even when all of the events are dissimilar, the novel occurrence of a retention interval may be sufficient to disrupt matching accuracy without affecting memory directly.
A potential solution to this problem is to expose pigeons to delays from the start of training. When this has been done, matching accuracy is found to be excellent over delay durations that would normally result in quite poor performance (Dorrance, Kaiser, & Zentall, 2000). Furthermore, when memory deficits are found they may look very different from those that have been reported when more traditional methods are used involving a small number of sessions with novel delays (Zentall, Klein, & Singer, 2004). In general, whenever the testing conditions are different from the training conditions, one should try to ensure that a decrement in performance cannot be attributed to the ambiguity of task instructions.
It may not always be obvious how to obtain an unbiased assessment of the ability of an animal, but a useful rule of thumb is to ask, what a human would do under similar circumstances. If the answer is not clear because of the absence of task instructions, one might not be able to obtain an unbiased estimate of the process being studied without modifying the task.
Such an approach that attempts to separate the mechanisms involved in task acquisition from appropriate understanding of the task instructions also has implications for research with humans. Typically, it is assumed that humans understand task instructions and that poor performance reflects a failure of memory or conceptual ability, however, this may not always be the case. When trying to interpret human performance, an awareness of the difference between these two kinds of failures may lead one to design better experiments with clearer instructions or independent assessment of comprehension of task instructions. Instructional failure is especially likely to occur when conducting research with adults who are developmentally delayed, with children, or with subjects whose culture is different from our own.
Dorrance, B. R., Kaiser, D. H., & Zentall, T. R. (2000). Event duration discrimination by pigeons: The choose-short effect may result from retention-test novelty. Animal Learning & Behavior, 28, 344-353.
Sherburne, L. M., Zentall, T. R., & Kaiser, D. H. (1998). Timing in pigeons: The choose-short effect may result from a confusion between delay and intertrial intervals. Psychonomic Bulletin & Review, 5, 516-522.
Zentall, T. R. (1970). Effects of context change on forgetting in rats. Journal of Experimental Psychology, 86, 440‑448.
Zentall, T. R., Klein, E. D., & Singer, R. A. (2004). Evidence for detection of one duration sample and default responding to others by pigeons may result from an artifact of retention-test ambiguity. Journal of Experimental Psychology: Animal Behavior Processes, 30, 129-134.