The researchers define the processes of seeking out the familiar versus the less familiar as exploitation and exploration, respectively. They begin with the concept of associations, using a free-association task. Basically, what this task involves is one person says a word, and the other says whatever comes to mind.
They examined word association among participants who were and were not under cognitive load. They manipulated cognitive load in different ways (across four experiments): digit-span task (memorize a certain number of digits), alphabetizing task (put letters from the word-association words in alphabetical order), and attending to features of the target words (color, shape). Overall, they found that responses were more diverse among participants under low cognitive load than participants under high cognitive load. That is, responses among people under high cognitive load were strongly associated with the target word.
So what overall conclusion can be drawn from these findings? As the authors state in their discussion:
Our findings support the notion that exploration is the default mode: The brain has a basic tendency to go beyond the nearest associations and activate unique ones instead when resources are available.This seems to fly in the face of the "cognitive miser" claim - that people prefer to save their cognitive resources and go through life on autopilot, unless something happens that requires them to think. Instead, it offers some support for something I said in the post linked above: we're cognitive neophiliacs - we're drawn to the new and will use our cognitive resources to understand that new thing. This can include that "outside the box" thinking that Baror and Bar observed in their study.