Cognitive Vulnerabilities in Memory Augmentation

Technological advances allow us to capture, store, and process multiple streams of near-continuous data, to be later used as cues to help retrieve associated episodic details. Our studies examine show that repeating such cues not only improves later memory for their captured events but could also lead participants to be vulnerable to forgetting of events related due to retrieval-induced forgetting.

Cinel, C., Cortis Mack, C., & Ward, G. (2018). Towards augmented human memory: Retrieval-induced forgetting and retrieval practice in an interactive, end-of-day review. Journal of Experimental Psychology: General, 147(5), 632.

We have extended these studies from real-life experiences to simulated presentations of experiences and advertising, and these studies are in preparation.

Detecting Inaccuracy in Augmented Memory

Retrieval-Induced Forgetting is just one of a number of cognitive vulnerabilities that are exposed through using technology. We have reviewed a range of such vulnerabilities, and present new data examining  further cognitive vulnerability is whether augmented memory data that is retrieved is inaccurate (by accident, technological failure, or design). We further examined whether providing inaccurate information for retrieval practice, can hinder memory performance for the other related encountered events. Particularly, we were interested in whether foils presented in the retrieval practice phase would be simply ignored, or whether participants assimilate them into their own streams and whether this will affect RIF. We also wanted to measure how sensitive participants are to inaccuracies in their photo stream and how this affects their trust in the used technologies

Clinch, S., Cortis Mack, C., Ward, G., & Steeds, M. (forthcoming). Technology-mediated memory impairments.. In Dingler T. & Nifaratos, E. (Eds). Technology-augmented Perception and Cognition. Springer.

Examining the Accuracy of Memory over time

More generally, how can consumers trust the integrity of the data captured and displayed?  What would make a client lose confidence that the data are authentic and veridical? Although technological solutions may be derived that seek to ensure the provenance of data, the trust and consent in using such solutions is ultimately limited by the users’ own abilities to detect inaccuracies in the re-presented memory cues.

One PACTMAN initiative has been to develop an iPhone app, RECAPP-XPR, that allows memory researchers to present experimentally-controlled stimuli over extended presentation schedules and then later test memory over long retention intervals. Conventionally, laboratory studies examining memory for word lists normally present and then test schedules lasting seconds or minutes. Using RECAPP-XPR, we have extended these data sets for schedules of words presented over hours and testing at the end of the day.

These studies confirm that memory accuracy is greater at the beginning and end of a schedule (primacy and recency effects, respectively), but these effects are much less pronounced than within conventional laboratory tests. However, there remain strong contiguity effects in recall –recalling one item leads to a heightened tendency to recall its temporal neighbours.

Cortis Mack, C., Harding, M., Davies, N., & Ward, G. (2019). RECAPP-XPR: A smartphone application for presenting and recalling experimentally controlled stimuli over longer timescales. Behavior Research Methods, 51(4), 1804-1823.