Table 3A.
Summary of compensatory and restorative studies (older adults and populations with health conditions).
| Article | Sample (completed) | Device | Active Control | Near transfer test | Far transfer* test | Retention test | Independent of company | Cognitive outcomes | Findings | |
|---|---|---|---|---|---|---|---|---|---|---|
| Ackerman et al., 2010 | 78 healthly adults (mean 60.7 yrs) | Wii Big Brain Academy | Reading exercises | Yes | No | No | Yes | Fluid and crystalline intelligence, processing speed | There was an effect of training on measures of processing speed (ps < 0.01) and fluid intelligence (verbal tests) (ps < 0.01), but there was no benefit for the TG compared to the control group for any measures. | |
| Åkerlund et al., 2013 | 47 participants with impaired WM following traumatic brain injury (47.7 ± 11.3 yrs) | Cogmed | No | Yes | Yes (SR) | No | Yes | STM, WM | Both TG and control improved on digit span (STM). The TG showed a significantly greater improvement in digit span (p = 0.045). TG did not show greater improvement on other measures of WM (spatial span, sequence memory). There was no self-reported change in executive function or psychological health for either group. | |
| Ballesteros et al., 2014 | 30 healthy older adults (57–80 yrs) | Lumosity | No | Yes | Yes (SR) | No | Yes | Processing speed, selective attention, EF, spatial working memory, episodic memory | TG showed greater improvement than controls in oddball task performance (selective attention), speed of processing and Wechsler memory (ps < 0.05), but not EF, spatial WM or self-reported wellbeing. | |
| Björkdahl et al., 2013 | 45 adults with WM deficits following brain injury (51.0 ± 11 yrs) | Cogmed | No | Yes | Yes | Yes | Yes | STM | TG showed significant improvement in STM (digit span) following training (p = 0.003), with no change in the control group. Both groups improved on therapist ratings of motor skill. | |
| Brehmer et al., 2011 | 23 healthy older adults (mean 63.7 yrs) | Cogmed | Non-adaptive version of training task | Yes | No | No | No | WM, sustained attention, inhibition, STM reasoning | Interaction effects indicated greater gains in divided attention and WM (span tasks) for the TG (p < 0.05). This was not found for inhibition, STM and reasoning. Training gains were related to changes in neural activation. | |
| Charvet et al., 2015 | 20 participants with cognitive impairment due to MS. (39.8 ± 11.5 yrs) | Lumosity | Computer games | Yes | Yes | No | Yes | Memory, processing speed | There was a significant difference between TG and active controls post-training in composite cognitive score (p = 0.02) but not for individual tests (e.g., WAIS, Corsi blocks). TG also performed better on a motor function task (p = 0.01). | |
| Edwards et al., 2013a | 74 adults (>40 yrs) with Parkinson's disease (68.9 ± 8.1 yrs) | InSight, Posit Science | No | Yes | Yes (SR) | No | No | Visual speed of processing | TG showed significantly greater improvements in visual processing speed (useful field of view) (p = 0.032). TG did not differ from controls in self-reported cognitive performance or depressive symptoms. | |
| Edwards et al., 2013b | 67 healthy older adults (74.0 ± 7.5 yrs) | InSight, Post Science | No | Yes | Yes (SR) | No | No | Visual speed of processing | TG showed significantly greater improvements in visual processing speed (useful field of view) (p = 0.043) than wait-list controls. There was no effect of training on self-reported social or cognitive function. | |
| Finn and McDonald, 2011 | 16 older adults with mild cognitive impairment (72.7 ± 7.1 yrs) | Lumosity | No | Yes | Yes (SR) | No | Yes | Sustained attention, WM, set shifting, visual memory | TG showed greater improvement on sustained attention, but not WM, memory (pattern recognition) or shifting (set shifting). Also no effect on subsequent training of waitlist controls. No effect on self-report of mood. | |
| Gropper et al., 2014 | 62 university students with ADHD or learning disabilities (28.0 ± 7.2 yrs) | Cogmed | No | Yes | Yes (SR) | Yes | Yes | WM, sustained attention, selective attention, reading and mathematics comprehension | There was no effect of training on WM (digit span), sustained or selective attention or mathematics and reading comprehension (ps > 0.05). There were similarly no group differences at 2-month follow-up. There was a reduction in self-reported ADHD symptoms. | |
| Haimov and Shatil, 2013 | 51 older adults with insomnia (65–85 yrs) | Cognifit | Simple computer tasks | Yes | Yes | No | No | Range of tests including: memory, divided attention, inhibition, shifting, WM, processing speed | TG showed improvements in several functions, including memory (p < 0.001), divided attention (p < 0.05), processing speed (p < 0.01), visual WM (p < 0.001). The TG showed greater improvements than the active control in memory (p < 0.001), visual (p < 0.001), and auditory (p < 0.001) WM. TG also showed improvements in sleep quality. | |
| Hellgren et al., 2015 | 48 adults with acquired brain injury (mean 43.7 yrs) | Cogmed | No | Yes | No | No | Yes | WM, processing speed, sustained attention, divided attention | The TG improved on all tests of WM and attention (ps < 0.001). TG reported increased quality of life (p < 0.001). No control group comparison. | |
| Hyer et al., 2016 | 68 older adults (>65 yrs) with memory impairment | Cogmed | Non-adaptive version of training task | Yes | No | Yes | Yes | WM, executive function | The TG showed greater improvements than active controls on one of two WM span tests (p = 0.01), but not on an executive function test. | |
| Kesler et al., 2013 | 41 women with history of breast cancer (56.0 ± 7.0 years). | Lumosity | No | Yes | No | No | Yes | Executive function, WM, processing speed | The TG showed significantly greater improvement than controls in EF (WCST) (p = 0.008) and processing speed (symbol search) (p = 0.009) but not WM (digit span) (p = 0.57). | |
| Klavora et al., 1995 | 10 participants (45–80 yrs) unsafe to drive following stroke | Dynavision | No | Yes | Yes | Yes | Yes | Processing speed | There was a significant improvement in performance on the trained task (ps < 0.05) and an increase in the proportion of participants rated as safe to drive. | |
| Legault and Faubert, 2012 | 41 healthy older adults (64–73 yrs, mean 66.3) | Neurotracker | Perceptual task (contrast detection) | No | Yes | No | No | None | At a distance of 4 m the TG showed significantly better perception of partially masked human motion than active controls (p = 0.040). There was no difference at 16 m. | |
| Leung et al., 2015 | 209 healthy older adults (70.1 ± 6.4 yrs) | Brain Fitness, Posit Science | Educational programme | Yes | No | No | Yes | Sustained attention, WM, verbal STM, | The TG showed greater improvement on one of two sustained attention tests (p = 0.026) and on working memory (p = 0.012) but not STM. | |
| Liu et al., 2016 (study 2) | 102 adults with ADHD (18–35 yrs) | Cogmed | No | Yes | No | No | Yes | WM, general intelligence | No effect of training group on changes in WM (delayed match to sample test) (ps > 0.05). | |
| Liu et al., 2017 | 88 young adults with ADHD (23.7 ± 3.3 yrs) | Cogmed | No | Yes | No | No | Yes | WM, fluid intelligence | No transfer to response control in Go/NoGo task. | |
| Lundqvist et al., 2010 | 21 adults with acquired brain injury (43.3 ± 9.8 yrs) | Cogmed | No | Yes | Yes (SR) | Yes | Yes | Divided attention, inhibition, switching, WM. | TG showed significant improvements in measures of WM, inhibition, switching, and divided attention immediately post-training (p < 0.001 to p = 0.002) and at 20-week follow-up (p < 0.001 to p = 0.002). There was no change in passive control group. TG also improved self-ratings of occupational performance. | |
| Mawjee et al., 2015 | 97 young adults (18–35 yrs) with ADHD (23.9 ± 3.4 yrs) | Cogmed | No | Yes | Yes (SR) | No | Yes | WM, STM, processing speed | There were no differences between TG and controls following training. There was also no difference in self-reported ADHD symptoms and cognitive failures. | |
| Mayas et al., 2014 | 27 healthy adults (57–77) | Lumosity | No | Yes | No | No | Yes | Alertness and distractibility | There was no effect of group (TG v control) on digit categorization performance in the oddball task. The TG significantly improved from pre to post in distractibility (p = 0.05) and alertness (p = 0.04). | |
| McDougall and House, 2012 | 41 healthy older adults (74.6 ± 8.5) | Nintendo Brain Age | No | Yes | No | No | Yes | Intelligence | Sub-tests of the WAIS only showed a benefit for backward digit span (p < .05). There was no effect for vocabulary, block design, arithmetic and forward digit span. There was no effect of more frequent use. | |
| Nouchi et al. (2012) | 28 healthy older adults (69.1 ± 2.4 yrs) | Nintendo Brain Age | Video game | Yes | No | No | Yes | Executive function, WM, processing speed | Following training, the TG showed significantly greater improvements in EF (ps = 0.001–0.006) and processing speed (ps = 0.005–0.014). There was no difference in WM (digit span) improvement between groups (p > 0.05) | |
| Peretz et al., 2011 | 155 healthy older adults (68 ± 7 yrs) | Cognifit | Video games | Yes | No | No | No | Overall cognitive performance | There was a significant improvement in overall cognitive score in the TG (p < 0.05) and the active control (p < 0.05). There was no difference in improvement between groups. | |
| Preiss et al., 2013 | 31 participants with unipolar and bipolar depression (44.2 ± 14.2 yrs) | Cognifit | No | Yes | Yes (SR) | No | No | WM, shifting, inhibition, divided attention, STM, executive function | There was no difference between TG and controls for WM, shifting, inhibition, divided attention, STM, or executive function (Stroop, WCST). Improvements in self-report of depressive symptoms. | |
| Rass et al., 2015 | 56 methadone maintenance patients (43.4 ± 8.0 yrs) | Cogmed | Non-adaptive version of training task | Yes | Yes (SR) | No | Yes | WM, STM, processing speed, reasoning, inhibition | Greater WM (digit span, OSPAN) improvements in the TG than controls (p = 0.003). No group differences in improvement in processing speed (trail making), inhibition or reasoning. TG reported less drug use post-training than active controls (p = 0.045). | |
| Siberski et al., 2015 | 32 adults with intellectual and developmental disabilities (40.5 ± 11.0 yrs) | Cognifit | Video games | Yes | No | No | No | Divided attention, inhibition, shifting, processing speed, WM | The TG improved in measures of monitoring (p = 0.017), visual WM (p = 0.003), and processing speed (p = 0.038) but not divided attention, shifting (WCST), or inhibition (Stroop). There were no group differences for any measure post-intervention. | |
| Smith et al., 2009 | 487 healthy older adults (>65 yrs) | Brain Fitness, Posit Science | Educational training | Yes | No | No | No | Cognitive assessment battery (inc. attention and memory), WM | The TG showed greater improvement than controls in the cognitive battery (p = 0.02) and WM task (p = 0.006). | |
| Strenziok et al., 2014 | 42 healthy older adults | Brain Fitness, Posit Science | Video games | Yes | No | No | Yes | Reasoning, WM, STM. | The TG showed a significant improvement in reasoning (WAIS matrix) scores (p < 0.05) but no improvement in WM (letter number sequencing). | |
| Von Ah et al., 2012 | 82 breast cancer survivors (56.5 ± 8.5 yrs) | InSight, Posit Science | Memory training | Yes | Yes (SR) | Yes | No | Memory, speed of processing | TG showed enhanced processing speed (useful field of view test) in comparison to passive controls post-training (p = 0.040) and at 2-month follow-up (p = 0.016). Also improved memory post-training (p = 0.0004) and at follow-up (p = 0.010). TG improved self-reported cognitive functioning (p = 0.042). | |
| Wentink et al., 2016 | 107 adults (45–75 years) recovering from stroke | Lumosity | No | Yes | Yes (SR) | Yes | Yes | WM, inhibition, fluid intelligence | TG outperformed controls in one of four WM tests (p = 0.02) and an inhibition test (p < 0.001) post-training. At 16-week follow-up there were no group differences in WM, inhibition, attention, and fluid intelligence tests. Also no differences in self-reported cognitive failures or quality of life. |
*
Transfer to tasks other than laboratory cognitive tests.
SR, self-report outcome; WM, working memory; STM, short-term memory; EF, executive function; TG, treatment group; OSPAN, Operation Span task; WCST, Wisconsin card Sorting Task; WAIS, Wechsler Adult Intelligence Scale; ADHD, Attention deficit hyperactivity disorder; MS, Multiple Sclerosis; ps, multiple p-values.
Quality assessment color key: