Event-Based Prospective Memory Deficit in Children with ADHD: Underlying Cognitive Factors and Association with Symptoms

Abstract

Event-based prospective memory (PM) was investigated in children with Attention deficit/hyperactivity disorder (ADHD), using a novel experimental procedure to evaluate the role of working memory (WM) load, attentional focus, and reward sensitivity. The study included 24 children with ADHD and 23 typically-developing controls. The experimental paradigm comprised one baseline condition (BC), only including an ongoing task, and four PM conditions, varying for targets: 1 Target (1T), 4 Targets (4T), Unfocal (UN), and Reward (RE). Children with ADHD were slower than controls on all PM tasks and less accurate on both ongoing and PM tasks on the 4T and UN conditions. Within the ADHD group, the accuracy in the RE condition did not differ from BC. A significant relationship between ADHD-related symptoms and reduced accuracy/higher speed in PM conditions (PM and ongoing trials), but not in BC, was detected. Our data provide insight on the adverse role of WM load and attentional focus and the positive influence of reward in the PM performance of children with ADHD. Moreover, the relation between PM and ADHD symptoms paves the road for PM as a promising neuropsychological marker for ADHD diagnosis and intervention.

Keywords: ADHD; attention; prospective memory; reward.

Figure 1

Procedure of Baseline condition (BC). The items consisted of 108 images of an animal located at the center of the screen; images of plants were located at the four corners of the screen. Participants had to press the “N” button when the image of a non-flying animal appeared on the screen and the “V” button when the image of a flying animal appeared on the screen (semantic decision task). Each stimulus appeared for 2 s. Participants received 1 RP for every correct answer and no RP for any wrong or omitted answer.

Figure 2

Procedure of 1 Target condition (1T). The items consisted of 120 images of an animal located at the center of the screen; images of plants were located at the four corners of the screen. As in the BC, participants had to press the “N” button when the image of a non-flying animal appeared on the screen and the “V” button when the image of a flying animal appeared on the screen (ongoing task). However, when a dove—PM target—appeared on the center of the screen (12 trials), participants were required to press the spacebar (PM task), instead of the semantic decision task. Each stimulus appeared for 2 s. They received 1 RP for every correct answer and no RP for any wrong or omitted answer, in both tasks (ongoing and PM trials).

Figure 3

Procedure of 4 Targets condition (4T). The items consisted of 120 images of an animal located at the center of the screen; images of plants were located at the four corners of the screen. As in the BC, participants had to press the “N” button when the image of a non-flying animal appeared on the screen and the “V” button when the image of a flying animal appeared on the screen (ongoing task). However, when one of the four PM targets (rabbit, crow, camel, pelican) appeared at the center of the screen (12 trials), participants were required to press the spacebar (PM task), instead of the semantic decision task. Each stimulus appeared for 2 s. They received 1 RP for every correct answer and no RP for any wrong or omitted answer, in both tasks (ongoing and PM trials).

Figure 4

Procedure of Unfocal condition (UN). The items consisted of 120 images of an animal located at the center of the screen; images of plants were located at the four corners of the screen. As in the BC, participants had to press the “N” button when the image of a non-flying animal appeared on the screen and the “V” button when the image of a flying animal appeared on the center of the screen (ongoing task). However, when the cactus plant—Prospective target—appeared in one of the four corners of the screen (12 trials), participants were required to press the spacebar (PM task), instead of the semantic decision task. Each stimulus appeared for 2 s. They received 1 RP for every correct answer and no RP for any wrong or omitted answer, in both tasks (ongoing and PM trials).

Figure 5

Procedure of Reward condition (RE). The items consisted of 120 images of an animal located at the center of the screen; images of plants were located at the four corners of the screen. As in the BC, participants had to press the “N” button when the image of a non-flying animal appeared on the screen and the “V” button when the image of a flying animal appeared on the screen (ongoing task). However, when a rhinoceros—PM target—appeared at the center of the screen (12 trials), participants were required to press the spacebar (PM task), instead of the semantic decision Table 2 seconds. In the ongoing task, they received 1 RP for every correct answer and lost 1 RP for any wrong or omitted answer. In the PM task, they received 50 RP for every correct PM answer and lost 20 RP for any wrong or omitted PM answer.

Figure 6

Accuracy on the ongoing task for each condition in the two groups. Within-group comparisons showed that in the CON group (A) no significant difference emerged between BC and all the PM conditions; the accuracy on the UN was lower than the RE condition. Conversely, participants in the ADHD group (B) showed lower accuracy in almost all the PM conditions compared to BC. 4T and UN conditions were lower than 1T and RE. Between-group comparisons (C) showed that the ADHD group performed significantly lower than the CON group on the 4T and UN conditions, while no difference emerged on the BC, 1T, and RE conditions. Asterisks mark significant differences: * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 7

Ongoing task speed for each condition in the two groups. Within-groups differences emerged in the mean response RTs (see A,B), but no significant differences occurred between groups (C). In the CON group (A), no difference emerged between the mean response RTs in the BC and almost all PM conditions (1T, 4T, RE), with the exception of the UN, in which they were slower than all other conditions. In the ADHD group (B), participants were slower in all PM conditions compared to BC; moreover, as in the CON group, on the UN they were slower than all other conditions. Finally, the mean response RTs on the 4T was slower than the RE condition. Asterisks mark significant differences: * p < 0.05, *** p < 0.001.

Figure 8

Accuracy on the PM trials for each PM condition in the two groups. Participants in the CON group showed no significant differences between conditions, with the exception of the UN, lower than the RE condition (A). Conversely, participants in the ADHD group showed lower accuracy on the 4T and UN conditions compared to 1T and RE (B). Moreover, differences between groups emerged because the ADHD group performed significantly lower than the CON group on the 4T and UN conditions (C). Asterisks mark significant differences: * p < 0.05, ** p < 0.01, *** p < 0.001.