How exactly blind individuals categorize and understand words relating to vision and sight has fascinated researchers and philosophers for hundreds of years. Seventeenth and eighteenth century thinkers like George Berkeley, David Hume, and John Locke were inspired by Molyneux’s problem, which asks what a blind person would know about the visual world if they became slighted gained sight? later in life. Many of these early thinkers concluded that a blind individual’s access to the visual world would be very limited and that they would need to relearn all aspects of the visual world once they were able to see. Similarly, early psychologists assumed that visual worlds were ‘meaningless’ to blind individuals and therefore their use of such words was just for social reasons.
In order to investigate this question, the current study by Marina Bedny and colleagues gathers ratings of semantic similarity between verbs from both blind and sighted individuals. Although this is not a perfect assessment of meaning, it can show us patterns of understanding that we can compare between these two populations. Similarity ratings have been used in the past to study blind individuals’ understanding of color by Shepard and Cooper in 1992. This technique allowed the current study to look at what details about visual verb meanings are consistent across blind and sighted individuals. They tested twenty-five congenitally blind and twenty-two sighted participants on 2041 (yes, two-thousand-forty-one) pairs of verbs each!
Surprisingly, judgements of visual verbs were indistinguishable between blind and sighted participants. In other words, both groups clustered meanings of verbs together in strikingly similar ways. Past experiments looking into similarity judgements for color words among blind individuals showed that there was a lot more variability in how blind participants understood color than how sighted participants did (Saysani et al., 2018; Marmor et al., 1978; Saysani et al., 1992). One theory about why this might be the case is that blind individuals pay attention to the most consequential pieces of information about the visual world and because the color of a strawberry may not be as important as whether someone glanced or glared, they learn those finer semantic distinctions about visual verbs as opposed to color.
Another interesting part of these similar clustering patterns between blind and sighted groups is that they revealed how accurately blind individuals represent aspects of visual verbs like temporal duration, intensity, and frequency. For example, they distinguish between glance and stare which indicates an understanding of how long each of those actions might last. They also distinguish between glow and twinkle which demonstrates an understanding of the high frequency of the latter. Finally, words like blaze and glow were also separated, presumably based on their respective intensities. Clearly, blind individuals are not only distinguishing between the context in which these verbs are used, but also between very complex aspects of their meaning.
One challenge a study like this faces is that simply rating the similarity between words does not reveal all the a person knows about a given word. Word meaning is heavily tied to the context around a given word and sometimes even the order of the words we hear can change the meanings we perceive. One example of this that I quite enjoy is “To say that surgeons are like butchers means something different than to say butchers are like surgeons” (Medin et al., 1993). Because of this context-dependent nature of meaning, the conclusions from this study are relatively limited. We gather that both blind and sighted individuals share common knowledge about the meanings of relevant “visual verbs” and that both groups pay similar attention to aspects of the domain of the word like modality or temporal duration when given these similarity-based tasks.