Enhancing Intelligence

Enhancing Intelligence

Richard Haier, a prominent intelligence researcher and the past president of the International Society for Intelligence Research, states, "In my view, the one ultimate goal [of intelligence research] is to find ways to increase intelligence — not [IQ] test scores, but the underlying intelligence factors, especially g". This view permeates his comprehensive The Neuroscience of Intelligence (2017 & 2023). In it, he concludes that, "despite many provocative claims and intriguing findings, no weight of evidence yet supports any means or methods for enhancing intelligence, fluid or otherwise." When test performance improves, the gains are hollow; they may improve IQ scores, but they do not change g, and they virtually all fade out over time (see The Fade-out Effect and how g is elastic rather than plastic due to evolutionary adaptations, and the Stability of g across the lifespan).

The primary obstacle to enhancing g is the hierarchical structure of mental abilities. In psychometric models, g sits at the apex, influencing performance on broader group factors and specific tasks below it. The arrows of causality in these models point downward, implying that training specific narrow abilities (at the bottom) is unlikely to evoke changes at the g level (at the apex). This is supported by studies showing that focal cortical lesions lead to local, not global (across all abilities), deficits, thus demonstrating that specific abilities are not causative of each other¹. This explains why (even if specific abilities are improved) behavioral interventions fail to show far transfer, i.e., the ability of a trained task to improve performance on unrelated cognitive demands.

Popular psychology has consistently promoted methods for enhancing intelligence that have failed to survive rigorous replication², including but not limited to: The Mozart Effect, n-back training and memory training, learning a language, compensatory education, Relational Frame Training (RFT or RRT), computer/video games in general, and drugs in general. More obscure but still undemonstrated is electrical stimulation of the brain, with four methods being researched: transcranial magnetic stimulation (TMS), transcranial direct current stimulation (tDCS), transcranial alternating current stimulation (tACS), and deep brain stimulation (DBS). In his book, Haier states:

From time to time, I am asked by magazine health writers to offer tips on increasing IQ. My answer is always the same and usually induces a long silence from the writer. There are no such tips – not even one that is supported by the weight of evidence. Eat better? Exercise? Engage in mentally challenging activity? All are good suggestions for general health and well-being but no specific effects for boosting intelligence can be substantiated. Not surprisingly, these writers never quote me, although science writers sometimes do in more substantial articles. I am happy to be the voice of reasonable skepticism to help stop the spread of bad information.

Richard Haier, The Neuroscience of Intelligence, p. 180².

Avenues

The evidence overwhelmingly points to the future of intelligence enhancement being in the province of the biological sciences. According to Haier, the two most likely research pathways toward enhancing intelligence, especially g, are (p. 181)²:

1. Neuroimaging studies that aim to identify specific networks and circuits in the brain. At some point, once the relevant ones are known, experiments to modify them could yield increases in reasoning and problem-solving ability.
2. Genetics research that gives clues about brain mechanisms at the level of neurons and synapses. This would open the door to experiments that modify the systems and their influence on reasoning and problem-solving.

Both pathways are complex and long-term projects, though advances in artificial intelligence may accelerate progress. The neuroimaging path may be shorter than genetic interventions, given current technology. But, in the long run, genetics may hold the key to the most dramatic enhancements possible. Several open-access papers on the Intelligence & Cognitive Abilities journal discuss such pathways. Advances in embryo selection may be of interest; while intelligence is highly polygenic, involving thousands of small-effect genes, researchers are investigating polygenic scores (PGS) for embryo selection to maximize the probability of high intelligence. Herasight boasts expected gains of 4 IQ points when three (the typical amount) embryos are available to choose from. Eventually, gene-editing technologies like CRISPR/Cas9 may be used to treat genetic disorders causing low intelligence, providing a roadmap for broader enhancement. Ultimately, the rapid convergence of psychometrics, genomics, and neuroimaging suggests that the goal of dramatically increasing human intelligence is no longer confined to science fiction.

References