Memory performance was lower for screenshotted than viewed material across 7 preregistered experiments, while no consistent compensatory benefits were observed across the memory and cognitive outcomes assessed, according to a study published in Memory & Cognition.
Investigators conducted 7 experiments involving 892 undergraduate students aged 17 to 27 years at a public university in the northeastern US. Each participant provided data to 1 experiment. In Experiments 1 and 4, demographic data were not collected. The experiments examined whether memory impairment associated with screenshotting was accompanied by concurrent or subsequent cognitive benefits.
Participants used their own smartphones to view or screenshot visual material before completing recognition, recall, source memory, or cognitive tasks. Experiments 1 through 5 used incidental encoding, whereas Experiments 6 and 7 examined conditions intended to encourage more conscious cognitive offloading. Designs included interleaved screenshot and viewing trials, blocked conditions, and between-group comparisons.
In Experiment 2, recognition accuracy was 70% for screenshotted images compared with 78% for viewed images. Raw source memory accuracy was 54% for screenshotted images compared with 71% for viewed images. Among correctly recognized items, source accuracy was 59% for screenshotted images compared with 70% for viewed images.
In Experiment 3, recognition accuracy was 61% in the screenshot condition compared with 88% in the viewing condition. Investigators detected no statistically significant difference between conditions in the number of modular arithmetic problems completed correctly.
Experiment 4 found no statistically significant difference in recognition of a subsequently viewed image set according to whether participants had screenshotted or viewed the preceding set. In Experiment 5, the preregistered analysis using the conservative recall measure detected no statistically significant difference in Block 2 recall. In an additional analysis using liberal scoring, recall of the second block was 19% among participants who screenshotted the first block compared with 16% among those who viewed both blocks. The additional analysis suggested a small advantage in subsequent recall under the more permissive scoring approach.
In Experiment 6, instructions informing participants of an upcoming memory test did not significantly modify the screenshot-related difference in recognition memory. In Experiment 7, recognition accuracy was 68% for screenshotted images compared with 75% for viewed images, with no interaction between screenshotting and participants’ expectation that the memory test would occur during the 30-minute session or approximately 1 month later. Participants were told they would be able to review their screenshots before testing, although in reality, all completed the memory test approximately 10 minutes after encoding without reviewing the images.
“The degree of ‘digital amnesia’ observed in these experiments is striking, especially given our remarkable capacity for image recognition,” wrote lead author Sophia P. Fabrizio of the Department of Psychology at Binghamton University, State University of New York, and colleagues.
The investigators stated that the findings may be most relevant to younger adults from a generation of digital natives and to digitally captured information that was not subsequently reviewed. They noted that additional research is needed in older adults and those with cognitive impairment.
“This suggests that the value of screenshotting as a memory strategy likely depends on how it is used,” wrote the investigators.
Disclosures: The authors reported no conflicts of interest.
Source: Memory & Cognition
