Researchers investigating the accuracy of eyewitness accounts staged and made a video of a crime, and showed it to test subjects.
Statements
The paragraph describes a Curious Difference between two groups, so it presumably wants us to extract the Causal Difference-Maker.
Since being told / not being told "the criminal might not be in the lineup" was the difference between the groups, and in one of them 78% misidentified the criminal whereas in the other one only 38% misidentified the criminal, we want to articulate that the 1st difference caused the 2nd difference.
Being warned that the suspect might not be in the group led to a lower rate of misidentifying the criminal.
Maybe they looked with a more critical eye, knowing it's possible that all the options might be wrong. Meanwhile the group who wasn't warned the criminal might not be in the lineup was more likely to "force it", to "make it work", to figure out which suspect looked closest and then convince themselves it was the criminal.
Goal
Look for something safely reinforcing the relationship that , or conversely, .