Each day, thousands of children struggle academically because they have difficulty maintaining their focus and attention.  This is true not only for children with ADHD, but also for those whose attention problems are more moderate, and may not rise to a level that warrants a formal ADHD diagnosis.  In fact, a number of published studies indicate that it is difficulties with attention, more than any other child behavior problem, that undermines children's academic success.

Given the adverse impact that attention problems have on children's academic performance, one would expect that extensive research on whether attention is skill that can be improved with practice has been conducted.  However, when my colleagues and I searched the literature several years ago for studies addressing this simple question, it was evident that very little work had been done. Those studies that we did find, however, reported promising finding which encouraged us to undertake a larger study that examined this important question [Rabiner et. al., (2009). A randomized trial of two promising interventions for students with attention difficulties. Journal of Abnormal Child Psychology.]

Participants were 77 1st graders from 5 elementary schools who had been identified as highly inattentive by their teacher, i.e., all students scored in the top 15% for the inattentive symptoms of ADHD on the Conners Teacher Rating Scale.  Very few of these students had been diagnosed with ADHD and our intent was to examine the impact of attention training in a diverse group of children with attention problems rather than focusing exclusively on a diagnosed sample.  Nearly 70% were male and the majority were either African American (58%) or Hispanic (24%).  This was a low-income sample and over two-thirds of the children qualified for free or reduced lunch.

Children were randomly assigned to 1 of 3 experimental conditions: Computerized Attention Training (CAT), Computer-Assisted Instruction (CAI), or wait-list control group (these children received intervention the following year.  A brief summary of the 2 types of computer interventions is provided below.


- Computerized Interventions -

* Computerized Attention Training (CAT) *

In CAT, the computer is used to present exercises where success depends on the ability to sustain attention and respond to clearly defined rules, e.g., pressing a certain key whenever a particular stimulus is flashed on the screen.  As the individual experiences success, the exercises become more challenging and the demands on attention increase.  Thus, structured practice in learning to attend is provided.

The CAT program used in this study was Captain's Log, a commercially available program that provides structured opportunities for exercising attention.  Ten exercises that focus on training auditory and visual sustained attention were selected.  As examples of the exercises used, one required children to press the space bar each time a symbol appeared that matched one already shown on the screen, and to refrain from responding to all other stimuli.  In another exercise, the children listed to a sequence of tones and had to decide whether a second sequence that followed was the same or different. 

To pass each exercise and move ahead in the program, children had to pay close attention and follow the rules about how to respond.  As they advanced, the exercises became more difficult and lasted longer.  Thus, they were practicing paying attention to tasks that became increasingly challenging over time.

* Computer-Assisted Instruction (CAT) *

CAI is another promising intervention for students with attention difficulties.  Unlike CAT, where successful performance requires sustaining attention to tasks that differ considerably from traditional academic activities, CAI involves presenting academic material via computer using instructional features that improve performance in children with short attention spans, e.g., providing immediate and frequent reinforcement and highlighting important information.  Because children must sustain attention to cognitive tasks of increasing difficulty, CAI may have attention training effects in addition to teaching basic academic knowledge and skills.

The CAI program used was Destination Reading and Destination Math published by Riverdeep.  At the 1st grade level, Destination Reading targets phonemic awareness, phonics, fluency, vocabulary, and comprehension.  Destination Math covers number sense, counting, addition and subtraction, comparing and ordering.  Basically, the program is similar to educational software that parents might purchase for their child, although it has sophisticated performance measuring and tracking capabilities.

- Intervention Implementation -

The computer training activities took place twice per week after school for 14 weeks.  Each session lasted 75 minutes during which children spent 50-60 minutes training on the computer.  Training took place in small groups of 5-7 students who were supervised by 2-3 research assistants.  To promote cooperation and motivation, each group could earn rewards, e.g., pizza party, for following rules and working hard.  Each child also earned stars for good performance on their computer training and could redeem these stars for small prizes.  Attendance was very good with children attending an average of 82% of scheduled sessions.  Most children received between 15-20 hours of actual training.


- Measures -

Ratings of children's attention were collected from their teacher before the intervention began, immediately after it ended, and 6 months later during the fall of the following year.  Academic achievement was measured at the same intervals using standardized achievement tests.  Teachers also completed ratings of children's academic performance in the classroom.


- Study Questions -

The study was designed to answer 3 basic questions:

1. Does either computer training activity result in children showing improved attention in the classroom?  This question was answered by comparing the change in attention ratings for children in each intervention condition with those in the wait-list control condition.

2. Does either computer training activity result in children showing improved academic achievement and performance in the classroom? This question was answered by comparing the change in academic achievement scores and teacher ratings of children's academic performance for children in each intervention condition with those in the wait-list control condition.

3. If any training benefits were evident immediately after training ended, do they persist into the following year?  This question was answered by comparing outcomes for treatment and control groups at the 2nd grade follow-up.


- Results -

Question 1 - Does training result in better attention in the classroom?

Both CAT and CAI were found to produce modest benefits in children's ability to attend in the classroom immediately following the training.  Specifically, there was a significant difference in the percentage of children in each training group and the control group who showed a moderate reduction in attention problems based on teacher ratings, which we defined as a reduction in attention difficulties of at least 0.5 standard deviations.  This reduction was attained by roughly 50% of intervention children (similar results for CAT and CAI) compared to only 16% of control children and was not related to whether teachers knew who was receiving intervention.

We also examined the percentage of children whose attention problem score was rated in the 'normal range' after training.  This was the case for nearly 25% of intervention children (similar results again for CAT and CAI) compared to only 4% of control children.


Question 2 - Does training lead to better academic achievement and teacher ratings of academic performance?

Findings here were more limited.  Children who received CAI were more likely to improve in reading fluency than control children.  No other affects on academic achievement were found.  Those in the CAI group were also more likely than control children to show improved teacher ratings of academic productivity, i.e., doing better at getting their work completed, and in their academic impulse control, i.e., approaching their work in a more deliberate and less impulsive manner.


Question 3 - Do beneficial effects found immediately after training persist into second grade?

Unfortunately, the answer to this question was no.  By the second grade follow-up, positive results for attention found at the immediately following the intervention were no longer evident.  This was not because children who received the intervention got 'worse' after it ended.  Instead, most children - including control children - showed a decline in attention problems according to ratings provided by their new teacher.

However, one interesting exception to the above was found for children who showed the highest level of attention problems to begin with, i.e., at least 6 symptoms rated at the highest level at the first assessment.  At the second grade follow-up, roughly 60% of intervention children in this most highly symptomatic group were reported to show either 0 or 1 symptom by their new teacher compared to only 20% of the control children.

Academic benefits that were evident for children who received CAI immediately after the intervention ended were also no longer evident in second grade.


- Summary and Implications -

Our primary objective was to test whether CAT or CAI helps inattentive first graders attend better in the classroom and perform better academically. Both interventions reduced inattentive behavior during first grade as roughly 50% of intervention participants showed declines in attention difficulties of at least 0.5 standard deviations compared to only 16% of controls. In addition, nearly 25% of intervention participants had inattentive scores in the normal range immediately following the intervention compared to only 4% of controls. Children receiving CAI also made gains in reading fluency and in ratings of academic functioning. 

Although these are encouraging results, fewer than 25% of participants received normalized ratings of attention difficulties at the end of first grade.  Furthermore, improvements in most standardized measures of academic achievement were not seen for either intervention and initial intervention effects for attention were not sustained into the next school year. As noted above, this was primarily because attention difficulties declined significantly for control children between 1st and 2nd grade. Thus, one explanation for the overall pattern of results is that the interventions accelerated what appears to be a strong normative decline in inattentive behavior during early elementary school. 

Although there was no evidence of long-term benefit in the main analyses, post-hoc comparisons conducted with the most inattentive children, i.e., participants who began the study with at least 6 inattentive symptoms, suggested a different conclusion. Among this subsample, participants were more likely than controls to have either 0 or 1 symptoms in 2nd grade, 64% vs. 22%.  This is an intriguing finding, but should be viewed cautiously given that this was a post-hoc analysis and the small sample size precluded the multi-level approach used for the primary analyses. 

Overall, the results provide initial evidence that CAT and CAI can improve children’s ability to attend in the classroom and suggest that benefits may be greater among children with particularly high numbers of inattentive symptoms.  To our knowledge, this is the first time that such an effect has been demonstrated. 

Given the modest magnitude of the effects on attention that were obtained, however, it is reasonable to ask whether these interventions can make meaningful contributions to the clinical care of young children with attention difficulties. Based on results from this study alone, the answer to this question may be no. However, rather than abandoning these approaches because the benefits were modest, the findings support the value of evaluating more intensive and/or different training paradigms to learn whether greater benefits than those obtained here may be possible.