Encouraging the Voluntary Mobilization of Mental Resources by Manipulating Task Design: Explorative Study

Cognitive training is increasingly being considered and proposed as a solution for several pathologies, particularly those associated with aging. However, trainees need to be willing to invest enough mental effort to succeed and make progress.

In this study, we explore how gamification in a narrative context (ie, the addition of visual game-like elements [GLEs] embedded in real-world contexts) could contribute increase in perceived playfulness (PP) and voluntary mental effort allocated to a cognitive task. In such context, narrative elements and GLEs can be designed to align with a commonly relatable scenario (like simulating fishing or gardening activity) to ground the task in familiar, real-world contexts. We also consider if the supposed effect of GLEs on PP and voluntary mental effort could endure while manipulating an intrinsic variable of the task (ie, by increasing cognitive solicitation).

In total, 20 participants (average age 33.6, SD 8.6 y) took part in 3 cognitive tasks proposed in a numerical format: a classic version of the Corsi test (Classic Corsi, a spatial memory task), a playful version of the Classic Corsi test (Playful Corsi), with added visual GLEs in a narrative context, and a playful version of the Classic Corsi test with added cognitive solicitation, that is, mental motor inhibition (Playful Corsi Multi). We assessed the impact of visual GLEs and cognitive solicitation on PP (1 question) and mental workload (MWL) using NASA-Task Load Index (NASA-TLX) and workload profile (WP) questionnaires.

Results showed that PP was not influenced by interface's playful characteristics (Classic Corsi [mean 62.4, SD 8.8] vs Playful Corsi [mean 66, SD 8.8]; W=77; P=.30) but decreased the time necessary to complete the task (Classic Corsi [mean 10.7, SD 2.1 s] vs Playful Corsi [mean 6.8, SD 1.6 s]; W=209; P<.001) as well as performance (Classic Corsi [mean 92.4, SD 9.1] vs Playful Corsi [mean 88.2, SD 11.3]; W=140.5; P=.02). So, possibly, visual GLEs could raise the stakes of the task slightly and implicitly encourage people to go a bit faster. Furthermore, visual GLEs increased MWL regarding attentional resources (assessed by WP: Classic Corsi [mean 52.4, SD 10.9] vs Playful Corsi [mean 65.8, SD 10.9]; W=27.5; P=.04), while manipulating cognitive solicitation impacted MWL when linked to task requirements (assessed by NASA-TLX: Playful Corsi [mean 54.2, SD 9.4] vs Playful Corsi Multi [mean 67.5, SD 9.4]; W=35.5; P=.01) without impacting the performance to the task (Playful Corsi [mean 83.8, SD 13.9] vs Playful Corsi Multi [mean 94, SD 5.5]; W=27; P=.007). Thus, working on the way cognitive functions are solicited would be wiser than adding visual GLEs to improve users' voluntary mental effort while preserving performance.

These results offer valuable insights to improve users' experience during gamified cognitive tasks and serious games.