Grades of evidence
‘Evidence-based’ brain training can mean many things. Cherry picking studies to back up claims for diets, health programs, and brain training is common place. Take virtually any claim and there will be some ‘evidence’ for it. The highest grade of evidence is the meta-analyses or systematic review of randomized controlled trials, followed by randomized control studies. Quality of evidence is high and risk of bias is low.
Working memory training
Here is a recent meta-analysis looking at the effectiveness of working memory (n-back) training (as implemented in apps such as i3 Mindware) in increasing fluid intelligence (Gf). The conclusion is that there significant improvements from training.
This article can be downloaded here.
The same level of effectiveness is also found in other meta-analyses that have been conducted (Soveri et al, 2017; Melby-Lervåg & Hulme, 2013; Weicker et al., 2016; Melby-Lervåg & Hulme, 2016; Melby-Lervåg et al., 2016.)
The scientific consensus has now converged on the conclusion that there are substantial cognitive skill gains from cognitive training that can be applied beyond the training game itself (called ‘near transfer’) and smaller general cognitive benefits to information processing not involving overlapping skills (called ‘far transfer’). This is reviewed in Appendix 1 below.
Proactive interference (PI) training
When you have PI, you have memory intrusions from prior memories that are no longer relevant to the current task you are focused on, and this can impair general cognitive performance.
PI is critical to higher cognitive functions. It contributes to short-term forgetting. It is closely associated with working memory capacity and higher cognitive functions, including general intelligence (G). It is also related to mental health and the ability to control negative ruminations.
There are numerous randomizied control studies that demonstrate the effectiveness of interference control brain training. A recent randomized control trial demonstrated that working memory (dual n-back) combined with interference control training results in additional improvements in both memory and language. This article can be accessed here.
This training effect may be because of the known link between working memory capacity and fluid intelligence (Gf): the abilities to control attention and inhibit interference account for the intelligence-working memory relation. This article can be accessed here.
Attention training
This recent meta-analysis of attention training as implemented in the IQ Mindware apps – especially gFOCUS – concludes that the training is effective, and generalizes to academic and cognitive skills. This review paper can be accessed here.
Tri-Network Framework for Brain Training
My own G CODE training is designed to target three main brain networks and their dynamic coordination in the service of comprehension, learning, problem solving, and strategizing, as well as general aerobic and autonomic health and adaptability. These networks (left) and their intelligence (G)-related function mappings (right) are shown here:
Boosting the Effectiveness of Brain Training: Brain Network Modularity
Multiple studies are now showing that a dynamic property of our brain networks called network modularity is a biomarker for how effective brain training will be in increasing cognitive ability. G CODE brain training is designed to increase the brain’s modularity through multiple strategies and thus improve training gains.
This recent systematic review of the evidence concludes that the more modular someone’s brain is, the more they will benefit from their brain training. The article can be accessed here.
Brain modularity and methods to increase modularity for brain training are explained in these two videos.
Heart Rate Variability (HRV) Training
The G CODE brain training system adopts Thayer and Lane’s (2009) well-known neurovisceral integration model. This proposes a neuronal network that relates heart rate activity and cognitive performance – with the central nervous system (CNS) CNS and autonomic nervous system (ANS) reciprocally interconnected such that information flows bidirectionally. HRV is a biomarker of vagus-nerve mediated parasympathetic ANS activity, and which has a close functional relationship cognitive flexibility in tasks involving attention, working memory and inhibitory control. In a number of randomized control trials it has been shown that individuals with high HRV perform better on tasks involving executive function and cognitive flexibility than those with low HRV.
A number of studies have demonstrated that low-paced diaphragmatic breathing (coherence breathing) improves heart rate variability (HRV), strengthens ANS efficiency and promotes behavioral or emotional self-regulation . The G CODE system employs a particularly effective type of coherence breathing in which there is a periodic holding of the out-breath before the next inhalation. This method has been shown to be more effective than traditional coherence-breathing cycles.
Evidence for all of the above claims underpinning the HRV training component of the G CODE system is found in these research articles: (1), (2), (3), (4).
Mindfulness Meditation Training
Scientific interest in meditation has significantly grown in the past years. Reviews consistently demonstrate the cognitive benefits of mindfulness meditation practices (in addition to benefits for anxiety and depression). Here is one comprehensive review by Chiesa and his lab that can be accessed here.
The G CODE system trains both focused attention (FA) meditation and open monitoring (OM) meditation.
FA meditation supports convergent thinking, the process of generating one possible solution to a particular problem.
OM meditation promotes divergent thinking, a style of thinking that allows many new ideas to be generated and insight-based problem solving.
The evidence for the dual convergent-divergent benefits can be accessed here.
Exercise & Aerobic Capacity Training
The most recent systematic review and meta-analysis of the effects of exercise on cognition is Ludyga and colleagues’ 2020 study which can be accessed here.
This study reviewed 80 randomized controlled trials and found a significant, general benefit of exercise on cognitive performance, which increased with longer durations of exercise. The effect was more pronounced in exercise involving coordination.
There is also evidence that exercise may be particularly beneficial in promoting executive function, cognitive resilience and stress-tolerance – critical to metabolic aspects of general intelligence (G). Supporting research articles include: (1), (2), (3).
Exercise <> Cognitive Training Synergy
The synergy reviewed below applies to either exercise combined with fluid intelligence (e.g. i3 Mindware) brain training or exercise combined with mindfulness meditation training.
Increased brain entropy & adaptability
There are multiple papers showing a synergy effect for combing aerobic exercise with cognitive training. Here is a recent randomized control study showing that exercise prior to brain training (on the same DNB working memory training as implemented in the IQ Mindware apps) results in greater generalization (transfer) to other fluid intelligence-demanding tasks. The researchers conclude: “our findings suggest that the addition of antecedent physical exercise to brain training regimen could enable wider, more robust improvements.” This paper can be accessed here.
Wang and colleagues also found an increase in complexity (entropy) of the EEG signal after the synergy training when performing a cognitively challenging task that wasn’t trained. This effect was not found after n-back training alone. Higher brain entropy is associated with higher levels of intelligence, as well as Based on the data, they conclude:
Neurogenesis
- The cognitive training should be between 1-2 weeks after the birth of the new cells.
- The training must be cognitively challenging. Easier tasks do not result in cell survival.
- The training must be successful – there is a strong positive correlation between how well we learn a task and the number of surviving cells.
- The training must involve sustained effortful / concentration, and that more effort results in more cell survival.
-
Regular, restful 7-8 hour sleep and non-directive meditation also benefits brain cell survival and the hard-wiring of hippocampal neural circuits. Sleep deprivation negatively impacts hippocampal cell survival. Even mild sleep restriction may interfere with the increase in neurogenesis that normally occurs with hippocampus-dependent learning.
Neuroscientist Professor Tracey Shors has looked extensively at the cognitive performance effects of combining aerobic exercise with mental training. She calls this MAP (Mental And Physical) training. And she finds a strong cross-training synergy effect from this kind of training.
Shors’ conclusion from this body of research is:
“Physical activity, especially aerobic exercise greatly increases the number of new neurons that are produced in the hippocampal formation. In contrast, mental training increases the numbers that survive, particularly when the training goals are challenging. Both manipulations can increase cognitive performance in the future.”
Curlik & Shors’ review article can be accessed here. The research behind the MAP training synergy can be found in these research articles: (1), (2), (3), (4), (5), (6), (7), (8), (9), (10).
HRV Training <> Cognitive Training Synergy Training
A higher HRV (particularly high frequency) is associated with increased cognitive flexibility and brain entropy – as reviewed in this article:
Cognitive flexibility refers to the ability to flexibly adapt processing to changing environmental information, to guide thought and behavior and to allow directed action toward a goal, as reviewed by Ionescu in this article.
Based on the same logic developed by Wang, Moreau and colleagues (2019) (reviewed above), the G CODE System combines HRV training with fluid intelligence and/or mindfulness training to augment training gains.
Nondirective Meditation & Sleep Training
Nondirective meditation is distinct from mindfulness meditations and includes Acem Meditation, Clinically Standardized Meditation, the Relaxation Response, Transcendental Meditation and yoga nidra. Nondirective meditation activates the default mode (crystallized intelligence/Gc) network, as reviewed in this article which can be accessed here.
This study shows that nondirective meditation results in more processing of memories and emotions than during concentrated meditation or basic relaxation.
Like sleep nondirective meditation also helps with memory consolidation as demonstrated in this randomized control trial, among others. The paper can be accessed here.
It helps activate the parasympathetic NS mediated relaxation response which works as a homeostatic counterbalance to the hormesis response. (1) A very short dose of nondirective meditation can positively influence stress, sleep, and well-being, as demonstrated in this article which can be accessed here.
The role of sleep in memory consolidation is well established – as reviewed here and here
APPENDIX 1
Cognitive Skill Training for Near Transfer
The most recent systematic review for working memory updating (WMU) training as implemented in i3 Mindware and DNB Pro is shown below, published in November 2020.
This review concludes:
The article can be accessed here.
What do these terms mean?
Control groups are the comparison groups in the studies who do not do the brain training.
A moderate effect size is substantial – similar or larger than the effect sizes of well-established interventions and drugs in the health services.
What is ‘near transfer’ and what is ‘far transfer’?
Near transfer means the cognitive skills (such as working memory updating) that are trained using the app, transfer to other contexts outside of the app game itself. Far transfer means the brain training transfers to general cognitive abilities that are unrelated (or not directly related) to the skills practiced in the brain training.
Here is another recent general review of working memory training by a lab from the MRC Cognition and Brain Sciences Unit, University of Cambridge that reaches the same conclusion:
As does this 2018 study:
Thus the argument can be made that effective cognitive training could train the following general-purpose cognitive skills
- working memory updating and maintenance
- interference control
- attention focus and inhibition
- attention monitoring and flexibility
These are all skills that are critical to fluid intelligence (Gf) and are expected to benefit Gf-demanding tasks with overlapping skill-demands.
Brain Basis of Near transfer
Systematic reviews, including the 2020 Pappa and colleagues review, consistently demonstrate that working memory training results in long-term neuroplaticity change in this Gf frontoparietal network – as well as connected subcortical striatal memory updating circuits.
The brain training results in more long-term neural efficiency in these networks. As the researchers in the latest 2020 review put it: “neural efficiency resulting from plastic changes induced by working memory updating training.”
Hormesis Effects for Far Transfer
The latest 2020 working memory training review finds compelling evidence that training effect sizes are significantly larger in studies comparing the training group to a passive control group than to an active control group doing another demanding cognitive task. Passive control groups don’t do any cognitively demanding task.
Why is this?
One explanation that goes beyond specific cognitive skills for near transfer benefits is based on the well-established principle of hormesis as widely researched by Mark Mattson. Hormesis is when ‘good’ stressors activate various genes resulting in better neural bioenergetics and neuro-protection. The effect is equivalent to the general health and bioenergetic capacity gains known to result from exercise, fasting and phytonutrients.
And the 2020 Pappa and colleagues’ review shows that with longer durations of cognitive training, there is more cognitive benefit compared to passive control groups but not active control groups. This makes sense if the underlying transfer principle is hormesis.
Mattson’s work can be accessed here.