At present, Developmental dyslexia (DD) is widely viewed as a neurobiological disorder (Valdois, 2010). The cognitive differences associated with DD were initially recognized because of lexical difficulties (Morgan, 1896). Later research has focused primarily on understanding the cognitive and neurophysiological processes that may explain these observed educational difficulties, particularly with reading and writing. Consequently, a literature examining the non-lexical effects experienced by individuals with DD is lacking (Tunmer and Greaney, 2010).

Developmental dyslexia is defined as “a disorder in children who, despite conventional classroom experience, fail to attain the language skills of reading, writing and spelling commensurate with their intellectual abilities” (World Federation of Neurology, 1968). Difficulty achieving a high level of reading and writing ability results from slow and inaccurate word recognition and spelling, problems that remain despite adequate instruction and intact sensory abilities (Peterson and Pennington, 2012).

This long-standing deficit-centric view provides an incomplete picture. Nearly forty years ago, Norman Geschwind noted an increasing number of studies suggesting that those with DD have superior talents in certain non-verbal skills that relate to art, architecture, engineering, and athletics. He was the first to highlight a likely evolutionary basis for the differences observed and, further, he suggested that when a relatively broad swath of a population exhibits a seemingly adverse condition, it is worth asking whether there might be some countervailing advantage at play (Geschwind, 1982). Decades on, researchers continue to ask similar questions, including Eide and Eide (2019), who noted: “[T]he question we need to be asking is not what’s wrong with the dyslexic brain, but what is dyslexic cognition for, what are these brains really built to do?.” Nevertheless, within academia there has been relatively little research or progress in understanding DD-associated abilities.

DD is understood to have a clear genetic basis (Valdois, 2010; Paracchini et al., 2016; see Erbeli et al., 2021b for a recent review). It is a complex trait involving multiple genes with twin studies indicating a heritability of at least 60% (Paracchini et al., 2016). Genetic influence on DD remains stable from adolescence to early adulthood, with the same genetic influences manifested across development from childhood to early adulthood (Wadsworth et al., 2007), suggesting that DD may not inherently be a disorder of development so much as a difficulty that is encountered during development.

The genetic basis is not the only good reason for taking an evolutionary perspective in attempting to understand DD-associated cognitive differences. DD also affects a large proportion of the population, between 5 and 20% (Badian, 1984; Wagner et al., 2020) and is universal and cross-cultural. Its cross-cultural nature suggests that the differences in cognition that underlie reading difficulties must have evolved by the emergence of behavioral modernity (ca. 150,000–50,000 BP).

The need to read and write is unlikely to have exerted any evolutionary selection pressure. Indeed, in the context of human existence, the technology of writing is a very recent invention (ca. 5350 BP) (Englund, 2004), with mass use among the general population only occurring over the last 100 years or so. It is the only example of a cultural invention for which we assume that difficulty in use relates to some kind of deficit: if someone does not show an aptitude for, say, accountancy or computer programming, we do not assume they have a neurobiological disorder.

Despite the recognition that there may be an evolutionary basis for DD (Geschwind, 1982; Stein, 2001), there has been no real attempt to explain DD-associated cognition from an evolutionary perspective until recently. A new evolutionary theory (Taylor et al., 2022) proposes that successful adaptation in humans arises from collaboration between individual members who are specialized in different but complementary neurocognitive search strategies (Taylor et al., 2022). This theory was developed to help explain DD-associated cognitive differences, by providing the theoretical understanding of why search specialization is likely to have evolved and its significance for understanding human adaptation and cultural evolution.

Here we reexamine DD-associated cognitive differences from the perspective of cognitive search – a theoretical approach applied to DD for the first time. Central to this is the understanding that many aspects of cognition can be viewed from the perspective of search, characterized by a trade-off between exploration and exploitation (Hills et al., 2015). We challenge the traditional view that the cognitive attributes of individuals with DD result from incorrect development. Instead, we propose that the features of this form of cognition were strongly selected for. It follows that what have been traditionally regarded as deficits are trade-offs, and that these are balanced by specialization and enhanced abilities in complementary areas of cognition.

This article does not intend to debate the existing evidence concerning DD nor understand why such individuals experience difficulties with writing technology. Rather, our aim is to reinterpret the existing evidence from the perspective of cognitive search and to understand the overall pattern of information processing in individuals with DD. Furthermore, the cognitive search paradigm provides a useful framework for generating hypotheses as to why these underlying cognitive differences may have evolved. When this new cognitive search perspective is combined with other lines of evidence, it is possible to see that the cognitive attributes identified in people with DD have a strong evolutionary imperative.

Throughout the history of dyslexia research, approaches have focused on identifying and remediating learning difficulties associated with dyslexia. Here, we outline key theories regarding the nature of DD-associated differences, including cognitive and biological theories.

In 1896, Pringle Morgan, a British physician, described a case of “congenital word-blindness” in an intelligent boy who had developmental difficulties with reading and spelling (Morgan, 1896). Whilst “word-blindness” had been described previously, Morgan was the first to postulate an underlying congenital problem (Kirby et al., 2020). Morgan (1896) assumed that “word-blindness” in his patients was caused by a visual processing deficit specific to words and spelling representations. This understanding remained the principle hypothesis on which dyslexia research was based over the next decades (e.g., Hinshelwood, 1917; Orton, 1925). However, DD readers were found to retain adequate perception of visual details, suggesting that there must be other underlying factors (Frith, 1978).

By the 1980s, the phonological deficit hypothesis emerged as the dominant theory of DD. The theory proposed that an underlying deficit in the representation, storage or retrieval of speech sounds impeded grapheme-phoneme mapping, resulted in difficulties in spelling and reading in alphabetic languages (e.g., Bradley and Bryant, 1978; Vellutino, 1979; Snowling, 1981; Stanovich, 1988; Brady and Shankweiler, 1991; Shaywitz and Shaywitz, 2005). Typical tasks where DD readers may struggle could include detecting whether words rhyme, deleting the initial or final phoneme, as examples. On a neurological level, differences in activation patterns in the perisylvian regions of the left hemisphere have been observed while participants were engaged in tasks involving phonological processing, inclusive of rhyme judgment, verbal working memory and pseudo-word reading (Démonet et al., 2004). While evidence supports the idea that phonological deficits contribute to the reading and spelling challenges of individuals with dyslexia, it cannot explain other differences associated with DD, such as differences in motor coordination.

A logical way to challenge the former approaches directly is to hypothesize that they are secondary to something else. Nicolson and Fawcett (1990) posited that, rather than being a specific difficulty with reading and writing, DD appears to be a more generalized learning difficulty: a difficulty with automatization, a skill that reduces the burden placed on working memory. This view holds that difficulties in acquiring and automatizing skills underlie the challenges observed over a range of areas, from fine motor skills to cognitive skills required to succeed in reading, writing, and mathematics. Given the perceived range of deficits associated with DD, including those involving motor skills and automatization, Nicolson and Fawcett identified the cortico-cerebellar circuit as a neurobiological basis of automaticity (Nicolson and Fawcett, 1990, 2007; Nicolson et al., 2001). Testing this across several studies, they found relative deficits in the awareness of time (Nicolson et al., 1995) and in postural stability and muscle tone: characteristics associated with cerebellar dysfunction (Fawcett and Nicolson, 1999). The delayed neural commitment theory further develops Nicolson and Fawcett’s automatization and cerebellar deficit hypothesis by considering the developmental process of constructing certain neural networks. The theory proposes that individuals with DD take longer to build and rebuild the neural networks that lie at the foundation of the kinds of skills discussed above, particularly those that are language-based (Nicolson and Fawcett, 2019).

The magnocellular deficit hypothesis provides another explanation of the deficits experienced in DD. Stein (2001, 2019) proposed that DD is “a hereditary temporal processing defect, associated with impaired magnocellular neuronal development” (Stein, 2018, 9). He outlined how temporal processing appears to be supported by magnocellular neurons, specialized in timing, that act in networks throughout the brain: in the cerebral cortex, hippocampus, cerebellum, and brainstem (Stein, 2018). These are involved in many functions including visual, auditory, touch, and proprioceptive systems. When magnocellular neurons are impaired, as is proposed to be the case in DD, several functions could be affected.

Until now, literature on the negative aspects of DD has dominated the field. Stein highlighted that the relatively high incidence of impaired development of the magnocellular system in individuals with DD “would not be so common unless there were compensating advantages for dyslexia” (Stein, 2001, 13). Like Stein, Nicolson and Fawcett have noted (Nicolson, 2014) that cognitive and neurophysiological theories regarding automatization may also help to explain some of the enhanced abilities observed in individuals with DD. They have suggested, for example, that the delay in automatization presents a trade-off in that conscious access to information is retained, making it easier to modify and integrate information and, in turn, to facilitate innovation (Nicolson, 2014).

While there are other approaches to DD that are not covered within the scope of this review, those discussed highlight some of the key approaches to visual and phonological deficit theories. Since they fail to identify a core deficit, key biological debates such as the automaticity/cerebellar and magnocellular approaches are also outlined. In general, however, the emphasis has been deficit-centric with few studies taking a more holistic approach that would incorporate both positive and negative differences experienced by individuals with DD.

Reflecting this, tools for identifying DD also tend to be deficit focused and may vary according to the approach of different practitioners. In general, the field has moved toward a multifactorial model of DD, whereby it is understood that most cases of DD cannot be explained by a single cognitive (or linguistic) “deficit” (Thompson et al., 2015). Relying on any one factor has been found to result in misclassification; for example, O’Brien and Yeatman (2021) found that relying on phonological processing measures alone lead to misclassification of DD in 30% of cases.

Multifactorial models recognize that a number of factors contribute to spelling, reading and writing difficulties experienced with DD. Some of the reported cognitive difficulties include rapid naming, phonological and morphological awareness, visual-orthographic knowledge, and verbal working memory (e.g., Shaywitz et al., 1999; Shaywitz and Shaywitz, 2005).

While most assessors consider a range of factors, assessments still vary between assessors and differences also exist between languages (Erbeli et al., 2021b). It is also important to note converging evidence from both twin studies and molecular genetic studies which indicate that reading ability lies on a continuum, with the cut-off point being arbitrary (Erbeli et al., 2021b). Taken together these factors may contribute to variation in reported forms of dyslexia and prevalence rates.

Although the above theories contribute to our understanding of DD-associated difficulties, the deficit-centered view does not tell the whole story. Observations that individuals with DD appear to exhibit countervailing advantages unexplained by current theories opens a new possibility: are there actual, one-sided deficits, or might the areas of difficulty be the downsides of trade-offs that exist, with the upsides being specialization and enhanced ability in other areas of cognition?

In order to determine whether individuals with DD are specialized to have a particular form of cognition, it is useful to consider whether the observed areas of enhanced ability share any fundamental pattern. In the rest of this section, we give a brief overview of such areas that are thought to be typical of people with DD. This is not a comprehensive or critical overview, as the lack of research on strengths precludes it. Rather, our aim is to point to areas of enhanced ability that have been proposed to exist by researchers and practitioners over the past four decades (e.g., Geschwind, 1982; West, 1997; Eide and Eide, 2011; Nicolson, 2014; Schneps, 2014).

Areas of enhanced ability that are consistently reported as being typical of people with DD include seeing the big picture, both literally and figuratively (e.g., von Károlyi, 2001; Schneps et al., 2012; Schneps, 2014), which involves a greater ability to reason in multiple dimensions (e.g., West, 1997; Eide and Eide, 2011). Eide and Eide (2011) have highlighted additional strengths related to seeing the bigger picture, such as the ability to detect and reason about complex systems, and to see connections between different perspectives and fields of knowledge, including the identification of patterns and analogies. They also observed that individuals with DD appear to have a heightened ability to simulate and make predictions about the future or about the unwitnessed past (Eide and Eide, 2011).

Individuals with DD have been proposed to exhibit greater ability in various areas of creativity. This has inspired several studies that have reported evidence for enhanced creative ability in a number of realms ranging from freeform drawing and other artistic objects (Cohn and Neumann, 1977) to literary creativity (Rack, 1981). Studies of creative ability also show evidence of a heightened ability to connect and carry out unusual combinations of ideas (Cancer et al., 2016), as well as heightened ability in tasks requiring novelty, insight, and more innovative styles of thinking (Everatt et al., 1999).

Practitioners have long observed that there appear to be high proportions of people with DD in professions and courses of study that rely on these abilities such as art and design, engineering, and entrepreneurship (see e.g., Geschwind, 1982; Martino and Winner, 1995; West, 1997; Newman and Sternberg, 2012).

In the realm of entrepreneurship, there has also been a growing interest in the apparently large numbers of entrepreneurs with DD (Alexander-Passe et al., 2021). A study of entrepreneurs in the United States found that 35% were dyslexic, with 22% being highly or extremely dyslexic (Logan, 2009).

In students enrolled in higher education, the incidence of DD is particularly high in creative subjects like arts and engineering. Wolff and Lundberg (2002) studied students enrolled at the University of Gothenburg, Sweden. They found that the prevalence of DD was significantly higher in students studying fine arts and photography compared with students studying economics and commercial law. At Central St Martins, University of the Arts London, United Kingdom, 75% of foundation year students had some form of DD (Steffert, 1996), and at the Royal College of Art, United Kingdom, 29% of students self-identified as having DD (RCA, 2015). In a study covering several United Kingdom universities across four degree disciplines (engineering, law, medicine, and dentistry), Lemon and Shah (2014) reported that self-identified DD in engineering was 28% compared with 5% in law. These self-reported figures are particularly high given that most people with DD do not get diagnosed (Aston et al., 2019). In these studies, it is tacitly assumed that admission to the degree programs ensured a high level of subject-specific talent, and the conclusion is therefore that higher education students with demonstrable skills in the arts and engineering are more likely to be dyslexic than students in non-creative subjects.

Approaches to explaining DD must account for both the difficulties and the enhanced abilities that are typical of people with DD. All the proposed strengths outlined above relate in some way to seeking out the unknown, often at the expense of exploiting known information. A useful framework for tying together these observations is cognitive search, which involves a trade-off between exploration–exploitation.

As noted above, search is fundamental to how we understand behavior, from cognitive control over a range of domains to the evolution of cognition. We review existing data on individuals with DD covering a range of different cognitive domains and modalities. We examine evidence for cognitive differences from the perspective of cognitive search as characterized by the exploration–exploitation trade-off. We consider external search (i.e., perceiving and selectively attending to information in the external environment), then internal search (i.e., searching for information in memory or using information from memory to search for solutions to problems), and finally, neurophysiological characteristics. Note that different terms are used across disciplines to refer to an emphasis on exploration or exploitation (see Table 1).

Supporting evidence varies greatly, depending on the area of cognition under study and its perceived relevance in understanding reading and writing difficulties. Nevertheless, regardless of the modality or specific terminology used, the recurring pattern that emerges supports the hypothesis that people with DD can be viewed as being specialized in explorative (global) search.

We have considered research on DD from the perspective of cognitive search as characterized by the exploration–exploitation trade-off. Since the differences observed in people with DD were initially observed because of difficulties found in reading and writing, studies to date have primarily focused on areas of cognition relevant to understanding these difficulties. Despite this focus, when taken as a whole, the evidence reviewed shows a pattern of stronger explorative search specialization (global search) for those with DD, with the corresponding trade-off in weaker exploitation (local search). Here, we consider additional lines of evidence that are relevant to this more specific proposal of search specialization.

People with DD are currently classed as having a neurobiological or neurodevelopmental disorder (Bishop and Rutter, 2008; Valdois, 2010; American Psychiatric Association, 2013; Thapar and Rutter, 2017); this implies that brain development is disrupted in some way, leading to abnormalities and a dysfunctional brain.

In the introduction, we outlined several factors that suggest this form of cognition may instead have an evolutionary basis. First, DD’s high prevalence within the population along with its high heritability imply that DD may provide advantages that complement its better-studied disadvantages. Second, writing is a very recent technology, and the need to read and write is unlikely to have exerted any significant evolutionary selection pressure. Finally, DD has been shown to be strongly polygenic in nature (Erbeli et al., 2021b). In combination with DD’s prevalence and heritability, these factors support the notion that, rather than representing a disorder, DD-associated cognitive differences were selected for and confer some kind of fitness advantage.

We then highlighted that the various DD-associated proposed strengths are all fundamentally related to exploration: global abstract and spatial reasoning, inventiveness, dynamic reasoning (the ability to simulate and make predictions about the future or about the unwitnessed past), and so on. These observations have been consistently highlighted in the literature by practitioners (e.g., Geschwind, 1982; West, 1997; Stein, 2001; Eide and Eide, 2011; Nicolson, 2014; Schneps, 2014). While some have argued that creative abilities in DD are coping strategies rather than inherent capacities, we have outlined multiple lines of evidence that show this is not the case.

The behavioral strengths that have been proposed are unified by the pattern of exploration. This is especially conspicuous considering that the various practitioners who noted these different strengths had neither an awareness nor expectation that DD would correlate with strength in exploration. Furthermore, this shared pattern is particularly striking since it is not immediately obvious that talents such as global spatial reasoning and dynamic reasoning share the same fundamental pattern, however this pattern does become clear through the lens of search. Similarly, incidence studies show particularly high numbers of people with DD in areas of study or work that require explorative ability, e.g., artists, designers, engineers, and entrepreneurs. It is this shared pattern which motivated the hypothesis that individuals with DD are specialized in explorative search. We investigated this by re-examining the extant cognitive psychology and neuroscience evidence from the perspective of search. A pattern of explorative specialization across domains and at all levels of analysis, strongly suggests that the higher-level explorative abilities that have been observed, emerge from these fundamental brain level differences. While more study is required in each area, collectively, an overall pattern of exploitative weakness and exploratory strength emerges.

We also considered a range of other perspectives to examine the possibility that humans (including those with DD) are individually specialized in search. On the one hand, the same explorative bias pattern for those with DD is found at all levels of analysis, including behavioral, cognitive (including internal and external search) and brain level differences. On the other hand, the existence of functional constraints that limit increased capability in search beyond a certain threshold help to explain why specialization may have evolved. The difficulty of optimizing search at the individual level is also reflected in the fact that other social species have also evolved to specialize in different search strategies. Another perspective considered was the alignment of the environmental pressures that shaped human evolution, specifically, “variability selection” (Potts, 1998) with the evidence for cognitive search specialization. The chronological alignment of environmental variability with the likely period at which human search specialization evolved was also discussed. A deeper explanation of the evolutionary theory applied to DD has been presented previously (Taylor et al., 2022). A wide range of evidence assembled from multiple, disparate disciplines points to the same pattern that both associates DD with an explorative bias and explains why this is likely to have evolved.

Taking all these factors into account, the assumption of pathology breaks down. We propose that, taken collectively, these various lines of evidence strongly indicate that individuals with DD do not have a disorder but instead, are specialized in explorative cognitive search.

In diagnosing DD as a disorder, the implicit assumption is that the problem or “deficit” exists within the individual. However, considering DD-associated cognition as a search specialization raises the possibility that the “problem” of DD exists in our cultural assumptions and systems. Our failure to recognize these differences as exploratory specialization can harm individuals with DD by subjecting them to social structures that limit opportunities or promote harm: a form of structural violence (Baber, 2017).

We propose that the cognitive differences observed in individuals with DD are not simply reflective of variation in the population. Rather, the strong clustering between exploratory traits and trade-offs suggest that these differences are part of a pattern of specialization and were selected for during human evolution. Nor do these cognitive differences reflect evolutionary mismatch, i.e., traits that evolved earlier and are now merely vestigial or maladaptive (Tooby and Cosmides, 1992; Li et al., 2018). Rather, we argue that the form of cognition represented by DD plays an essential role in enabling humans to adapt.

Given the high prevalence of DD, this in turn would indicate that humans balance the trade-off between exploration and exploitation through specialization in complementary search strategies. Collaboration between these different strategies would allow more efficient and effective search in the co-creation of cultural adaptations, helping to explain our species’ exceptional adaptiveness. Just as genetic search mediates biological evolution, it has been proposed that complementary cognitive search mediates cultural evolution in humans (Taylor et al., 2022).

It is worth emphasizing that once a system’s components specialize, all of them become interdependent and thus essential. De-emphasizing those parts geared toward exploration tips the system excessively toward refining existing solutions. Cultural change may remain, but it is likely to become progressively less adaptive. Systems that refine existing solutions more rapidly than exploring new ones may be effective in the short-term but are self-destructive in the long-term (March, 1991).

Nurturing different individual cognitive strengths and fostering collaboration would help to realize the synergistic benefits of complementary cognitive search strategies. Removing obstacles to explorative learning, and instead harnessing exploration to increase adaptiveness may enable us to better confront the existential challenges presently facing our species and our planet.

The original contributions presented in this study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

HT: conceived the theory. HT and MV wrote the manuscript and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.