CogniFit - Computer-Based Cognitive Training for Individuals With Intellectual and Developmental Disabilities: Pilot Study

Original Name: Computer-Based Cognitive Training for Individuals With Intellectual and Developmental Disabilities: Pilot Study.

Authors: James Siberski1, Evelyn Shatil2, Carol Siberski3, Margie Eckroth-Bucher4, Aubrey French5, Sara Horton6, Rachel F. Loefflad7 y Phillip Rouse8.

• 1. Assistant Professor Gerontology, Misericordia Univesristy, Dallas, PA, USA.
• 2. Chief Scientist, Head of Cognitive Science, CogniFit Inc, Haifa, Israel.
• 3. Private practice Geriatric Care Manager, Nanticoke, PA, USA.
• 4. Professor Department of Nursing, Bloomsburg University, Bloomsbur, PA, USA.
• 5. Program specialist at the Institute for Human Resources and Sevices, Kingston, PA, USA.
• 6. Social Worker, Long Term Care, West Wyoming, PA, USA.
• 7. Recreational Therapist, Dallas, PA, USA.
• 8. Private Practice License Psychologist, Sunbury, PA, Usa.

Journal: American Journal of Alzheimer’s Disease & Other Dementias (2015), vol. 30 (1): 41-48.

References to this article (APA style):

• Siberski, J., Shatil, E., Siberski, C., Eckroth-Bucher, M., French, A., Horton, S., Loefflad, R.F., & Rouse, P. (2015). Computer-Based Cognitive Training for Individuals With Intellectual and Developmental Disabilities: Pilot Study. Am J Alzheimers Dis Other Demen, 30 (1), pp.41-48.

Study Conclusion

CogniFit personalized cognitive training is a promising program to improve cognitive performance in adults with intellectual or developmental disabilities through 20-30 minutes daily, 3 days a week, for 10 weeks. Visual memory, p=. 003; Planning, p=. 009; Monitoring, p=. 017 (Yes α=.05 or α=.1).

Study Summary

This study wanted to test how personalized cognitive training affects the cognitive abilities of adults with intellectual disability (ID) or developmental disability (DD).

Fifteen cognitive abilities were evaluated in 32 people with DI/DD. The participants were divided into three groups:

• 11 people in Computer CogniFit
• 11 people in control group with video games.
• 10 people in control group of the waiting list or daily activities.

Despite the high adherence to the study, the number of participants was very small and the results (both intra and inter-group) were not significant when applying a high level of α (α=.001). However, by applying slightly more permissive levels of α, 11 of the 15 cognitive abilities showed an important improvement in the computerized cognitive training group. It is estimated that, with a representative sample, the data could reach statistically significant levels.

These data are promising and could be a good basis for a future study using a larger sample. If confirmed, this would demonstrate the importance of people with ID/DD performing computerized cognitive training, such as that offered by CogniFit.

Context

Intellectual disability and developmental disability (such as autism or cerebral palsy) entail a number of very heterogeneous difficulties that translate into a deficit in adaptive behavior with respect to the rest of the population. In the United States, it is estimated that 5% of children in primary or secondary school have some type of ID/DD.

In the case of adults, information is scarcer, but the data suggest that percentage of adults with intellectual disabilities is 0.5 %, while percentage of adults with developmental disabilities is 0.4%. These disabilities can have a significant personal, family and social impact.

There is now a growing interest in improving cognitive skills through computerized tools, such as CogniFit. In spite of this, it is common for these people not to have easy access to these tools. This is often given by the erroneous belief that they are not able to use a computer.

It is important to change this erroneous belief because people with ID/DD have an equal or greater risk of suffering some type of cognitive impairment. This posed need to prove the benefit of computerized cognitive training for people with intellectual or developmental disabilities.

Methodology

Participants

Thirty-two participants were selected from among 145 adults with an intellectual disability or developmental disability who attended a specific rehabilitation center for people with these problems.The study was conducted in this same center, located in a city in northeastern Pennsylvania, USA.

The inclusion criteria to participate in the study were:

• Have a diagnosis of ID/DD
• Being over 21 years of age.
• Good health.
• Workshop employee.
• Wanting to participate in the study.

Exclusion criteria included:

• Lack of interest in computers.
• Inability to sit at a computer for 25-30 minutes.
• Inability to see images clearly on the computer screen.

Participants were told the objectives and requirements of the study. The participants' guardians signed the consent form. The board of Misericordia University approved the procedure for this study.

Design

A blind, randomized design was applied. Participants were randomly assigned to one of three groups:

• Computerized Cognitive Training Group (with CogniFit).
• Video game control group.
• Waiting list control group.

Participants were unaware of which group they belonged to, but the researchers did possess this information. In fact, the activities of the video game control group were designed to resemble those of the cognitive training group. The duration of the sessions, pre- and post-treatment evaluation were the same in both groups.

Process

The TONI-3 (Test of Non-verbal Intelligence, third edition) was used to evaluate the global intelligence level of each one of the participants.

A neuropsychological evaluation using CogniFit cognitive assessment. To avoid adaptation, the CogniFit tool automatically applies different versions of stimuli and tasks for each evaluation. It has also been scientifically validated. This assessment consisted of 15 tasks measuring 15 cognitive skills.

Finally, interventions were carried out in the rehabilitation center but in different rooms each. The interventions lasted 10 weeks and consisted of 3 weekly sessions of 20-30 minutes each one.

Control Group with videogames

Participants in this group did activities designed especially for this study. These activities were intended to be similar in appearance or duration to those of CogniFit, but did not have some of its most relevant characteristics for cognitive training: they did not fit the level of the user, did not require rapid response or attending to more than one stimulus at the same time in any of its tasks. Therefore, it can be said that it acted as a placebo, although they still allowed a certain degree of training.

Control Group with waiting list

This group maintained its usual schedule and activities. These consisted mostly of workshops, eating in a common dining room, having morning and evening breaks, etc.

Analysis

Statistical analyses were conducted through SPSS 17. To measure intragroup and intergroup differences in the 15 cognitive abilities studied, mixed effect models (with fixed and random effects) were used for repeated measurements, with a separate model for each variable.

Results and conclusions

The results showed that there were no statistically significant differences between cognitive ability scores in pre-training assessment and post-training scores. In fact, none of the 15 intragroup or intergroup comparisons were significant (p>. 001). Therefore, the cognitive training group did not show any relevant improvement in their cognitive abilities after training, nor did they show any relevant improvement with respect to the control groups (who also showed no improvement).

However, when criteria are less strict (p < .05 y p < .1), the experimental group shows significant effects in 11 out of the 15 cognitive abilities, in regards to the initial evaluation (Visual memory, p=. 003; Planning, p=. 009; Monitoring, p=. 017). Applying these criteria, there were also significant effects on 4 of the 15 cognitive abilities in the control groups. This data is very interesting for this type of study, especially when such a small sample has been used (which may contribute to the non-significant results at such a strict level). Through Cohen's d, it was observed that the effect size between the posts and the pretest in the experimental group was:

• Medium (Cohen's d between. 58 and. 72) in the following cognitive abilities: monitoring, planning, recognition, response time, cognitive flexibility, spatial perception, visual memory and visual scanning.
• Small (Cohen's d of .44, .37 y .41) in naming, visual perception, and working memory.

Therefore, the results of this study are not significant, possibly due to the small size of the sample. This limitation makes the results of this pilot study unrepresentative. Despite this, the data is encouraging. It is to be hoped that, in carrying out this same experiment with a larger sample, the results will be significant. In addition, 11 of the cognitive abilities studied are significant when applying less restrictive criteria (P <. 05 and P <. 1). This leads us to think that CogniFit can be a good tool for training cognitive abilities of people with intellectual disabilities or developmental disabilities.