Heit, E., Brockdorff, N., & Lamberts, K. (2004). Strategic processes in false recognition memory. Psychonomic Bulletin & Review, 11, 380-386.
The goal of the present article was to determine the extent to which false recognition could be avoided when people use strategic processing. This was examined in two ways: by examining the time course of the false memory effect, and by giving different instructions to different groups of participants.
Experiment 1
There were three hypotheses for experiment 1. 1. With standard instructions, there should be more of a false memory effect relatively early in the time course of making a judgment and 2. strategic processing should lead to a reduction of the effect. 3. Giving forewarning instructions should lead to a reduction of the effect.
Each study list consisted of Eight Nine-item DRM lists (72 words) that were presented at a rate of 3 sec per word. After each study list, participants were given a recognition test. The word would appear, and at variable time lags (200, 400, 600, 1,100 msec) a tone would sound, where participants were instructed to respond immediately. This happened four times. Participants given forewarning instructions were told of the nature of DRM lists and were given the king list as an example.
Results
Under standard instructions, after the first response signal, old responses to old items tended to increase, but there was a fall in old responses to critical lures. Under forewarning instructions, the proportion of old responses was lower for all three kinds of stimuli. There was, however, a greater effect of forewarning at earlier response signals. When the raw scores were converted to d’ units, they found that participants’ ability to distinguish between targets and critical lures was better at later response signals. The authors conclude that the results are consistent with the idea that the false memory effect is reduced by strategic processing.
Experiment 2
Another condition was added, one in which participants were given “inclusion instructions” where they were instructed to say “old” to critical lures as well as targets. The hypothesis coming out of this additional condition was that participants would be less likely to engage in strategic processing and so the reduction of the effect at later response signals (that was found in Experiment 1) would not be as evident in this condition.
Each study list consisted of 36 DRM items (nine items from 4 DRM lists) and 6 filler stimuli. After each list, participants completed the recognition test. This happened eight times.
Results
19% of the responses were trimmed due to early or late responding. The results for the standard and forewarning condition were similar to Experiment 1. The inclusion condition results were a lot different, with critical lure items receiving more old responses than the actual targets. The results fit all of their predictions, including the fourth – that participants could “avoid the strategic processing that would otherwise reduce the false memory effect.” p 385.
In conclusion, the authors suggest that their results show that the false recognition effect is strongest at early points during the time course of making a judgment, and weaker at later points. They go on to suggest that the false memory effect is therefore due to automatic processing.