This study concerns the effects of list repetition on false recognition in the DRM paradigm.  As a reminder in the DRM paradigm participants are presented with lists of words (e.g. table, sit, legs, seat, couch, desk, recliner, sofa, wood, cushion, swivel, stool, sitting, rocking, bench) and then falsely remember the related critical lure (e.g. chair).

Benjamin discusses two theoretical perspectives on the effects of repetition on false recognition.  First, according to the source monitoring perspective subjects generate the critical lure at study and then make a source monitoring error at time of test.  Benjamin argues that this account should predict that repetition would increase false memories because you would be increasing the opportunity to activate the critical lure at study.  On the other hand, the mirror effect is the general finding that manipulations that tend to increase target recognition decrease lure recognition.  The mirror effect thus predicts fewer false memories with repetition.

Benjamin's own view is that repetition increases the familiarity of the critical lure (perhaps through spreading activation) increasing the probability of false recognition.  But repetition should also increase the conscious recollection of targets which may cause subjects to set a higher decision criterion (see Dodson & Schacter, 2001; Israel & Schacter, 1997; Schacter, Israel & Racine, 1999).  So repetition should have effects that both increase and decrease false memories.  In these experiments Benjamin is trying to tease those two effects apart.

This experiment looks at aging.  The idea is this: The process that leads to false recognition (i.e. familiarity) is an automatic process that is preserved with aging.  The process that reduces false recognition requires recollection of targets, a process that declines with aging.

So there should be an interaction between age and the effect of repetition.  Repetition should increase false memories in the elderly but should decrease false memories with younger participants.

Subjects were presented with 10 DRM lists made up of 8 items each.  Subjects studied 5 of the lists once and 5 of the lists 3 times.  When a list was presented three times it was presented three times in a row without any intervening lists and lists alternated between lists that were presented once and lists that were presented three times.

The test consisted of 80 items

• Forty Targets: Four randomly chosen words from each list
• Ten Critical Lures: The critical lure from each of the ten lists
• Thirty Categorical Distractors: Three of these from each list.  These are what we would call missing exemplars, they are non-presented targets from the DRM lists

• Repetition increased true recognition for both age groups.
• Recognition of the critical lures decreased with repetition for the younger group but increased with repetition for the older group.
• There was no effect of repetition on the categorical distractors

Another factor that should influence conscious recollection is a response deadline.  Familiarity is thought to be automatic and quick while recollection slow and deliberate.  If Benjamin's account is right repetition should increase false recognition when subjects have to respond quickly but decrease false recognition when subjects are allowed to take their time.

The basic design was similar to Experiment 1 (a few more lists were used).  On the test subjects were presented with 9 item blocks containing a critical lure, four targets and four categorical distractors.  On half the blocks subjects were tested under time pressure by requiring them to respond in 750msec.  On the other half  of the blocks there was no deadline given.

Results

• Repetition increased target recognition under both speeded and unspeeded conditions
• Repetition increased false recognition under speeded conditions, but decreased false recognition under unspeeded conditions (although this was only significant if a one tailed test was used).
• Repetition had no effect at all on recognition of the categorical distractors

• Benjamin also reports some signal detection analyses showing that
• In Experiment 1, presenting items three times increased the ability of younger subjects to distinguish between targets and critical lures but had the opposite effect for older participants
• In Experiment 2, presenting items three times increased subjects ability to distinguish between targets and critical lures when there was no time pressure to respond but appeared to have no effect when there was time pressure.