When I was a kid, my parents needed to find creative ways to keep me busy, or else I would be a cranky bored kid. For a while, they kept me occupied with word search games where you get a big grid of letters and have to find the hidden words in them. At the end of the book, there would be the answers to each puzzle. Sometimes, I would just give up on a puzzle and check out the answer.
Does having access to the answers to a question change the effort that people put in to solve a problem? This is not just an idle question. Textbooks also often put the answers to selected questions in the back of the book. Understanding the effect of having access to answers has implications for education, then, as well.
This question was explored in a series of studies published in the December 2017 issue of the Journal of Experimental Psychology: General by Evan Risko, Michelle Huh, David McLean, and Amanda Ferguson. The results are interesting, but they also show how hard it can be to give a clear answer to a seemingly simple question.
The set up for these studies was simple. Participants were given a set of 5-letter anagrams to solve. Anagrams are scrambled groups of letters like SRNEA, which can be turned into words. This anagram has multiple solutions like NEARS, SNARE, EARNS, and SANER. Participants were either told that they would get feedback about the correct answers immediately after working on each anagram (and then received feedback) or they were not given feedback. The researchers were interested in how much time participants spent on the anagrams and how many of the correct answers they got.
Across all of the studies in this paper, participants generally spent more time-solving anagrams when they were not going to get feedback about whether they were correct than when they knew they were going to get feedback. So, knowing that they were going to get the answers soon led to less time on the task. Interestingly, though, people got about the same proportion of correct responses in each condition (getting about 60 percent right).
Why did the time differ?
In several of the studies, participants were also asked at the end of each anagram whether they were confident they had gotten all of the answers correct and whether they felt that if they had more time they would get more answers. In general, participants were moderately confident that they had gotten all of the answers correct (averaging about 60 percent confident) and felt that they were unlikely to get more answers correct if they took more time. This was true regardless of whether they had access to the correct answers.
At first glance, this pattern of results seems to suggest that having access to the correct answers decreases persistence on a moderately difficult task. That would seem to suggest caution about providing answers to problems for students.
But, notice that participants got about the same proportion of answers correct in each condition, even though the time spent on the task differed. In some studies, participants spent about 10 seconds more on average per problem when they did not get the answers than when they did. These 10 additional seconds per problem did not improve their overall performance. So, perhaps the people who quit earlier were actually doing something more optimal here because they did not waste as much time.
A reason to think this might be the case is that in one of the studies, participants did two blocks of trials. Participants who were going to receive answers only took less time on the anagrams in the first block of trials and not the second. This suggests that eventually all participants figured out how much time they needed to complete the task, and at that point having available answers did not affect their performance.
It may be that participants are actually using the answers to learn how well they are doing the task. One thing everyone must learn is to put in the right amount of effort on the problems they solve. Too little, and they could have been more accurate. Too much, and they are wasting time that could have been spent on something else. When participants feel like they are just about done and have access to answers, they might just check the answers to see how well they did. While those participants who did not have access to answers might spend more time just to be sure there weren't any additional solutions to the anagram that they missed. More research will be needed to figure that out.
This set of studies addresses an interesting and important topic, in a straightforward and clever way. The results are consistent and are obtained repeatedly. And yet, despite that, the studies still raise as many questions as they answer. That is part of the joy of science.
Risko, E.F., Huh, M., McLean, D., & Ferguson, A.M. (2017). On the prospect of knowing: Providing solutions can reduce persistence. Journal of Experimental Psychology: General, 146(12), 1677-1693.