Jou, J., Matus,
Y. E., Aldridge, J. W.,
Past literature has suggested two ideas of the differences in veridical (true) and false recognition that are at odds with each other. The first claims that false memories can be indistinguishable from true memories, while the opposition claims fundamental and detectable differences between true and false memories.
Experiment 1
This experiment was used to test for differences in reaction speed to targets and critical nonpresented words. An additional manipulation of presentation frequency was included to test for any potential changes in false memory strength.
Results and Discussion
Overall the RT for false alarms was significantly greater than the RT for hits to critical words. Additionally, there was a interaction between response category and the frequency of presentation, with RT decreasing as frequency increased for hits and RT increasing as frequency increased for false alarms. Analogous findings were reported for comparisons between critical nonpresented words and list items.
These data suggest that increased frequency of presentation leads to increased accuracy rates and faster response times. Additionally, the RT data shows that responses are faster for correct responses than false responses, i.e. faster responses for hit than for false alarms and faster responses for correct rejection than for misses.
Experiment 2
The main goal of experiment 2 was to test for any knowledge on the part of the participant of the differences between accurate and false recognitions.
Results and Discussion
Results of the ANOVA for critical words replicated the results of the same analysis from experiment 1. Also, as in experiment 1, RT was lower for hits than to false alarms. Again, as in experiment 1, there was a frequency by response type interaction.
Accuracy data showed significant differences in the false alarm rate of the three presentation condition vs. the two and one presentation conditions, but not between the latter two.
Confidence data was collected which significantly correlated (negatively) with RT. Further analyses suggested that confidence ratings were significantly different for each of the response types, with hits receiving the highest confidence ratings.
Experiment 3
Since the false memory rate were lower in the first two studies than what has been found in the past literature this third experiment was conducted to test for any differences in the types of results found in experiments 1 and 2 that might occur with lists more akin to those in the past literature.
Results and Discussion
Dealing with reaction time data, results of this experiment replicated the main findings of experiments 1 and 2, suggesting that list composition was not a contributing factor to the phenomenon.
General Discussion
Overall the data suggest that true and false recognition are distinguishable through reaction time data. These results concur with past results found in the physiological literature. Additionally with the negative correlation found between confidence and RT, there appears to be no dissociation between the two measures.
In closing the authors suggest that past reports of vivid false memories that have led researchers to suggest these false memories may be indistinguishable from true memory may, in fact, be due to demand characteristics caused by verbal report.