Binary classification as a cognitive strategy?

Question

Is there any research investigating whether the human cognitive system has a tendency to reduce complex systems or spectra of data in terms of binary contrast?

There are many common-sense dual taxonomical sets. Some of these have a basis in objective reality (the binary male/female has proved an especially effective evolutionary strategy), others may be argued to be more in the eye of the beholder. Day and night could be said to be extremes in a cycle; left and right, conservative and liberal are ranges on a spectrum (cf. friend and foe; "there are only two kinds of people"). A lot has been written about gender identity and sexual orientation in the framework of a Hegelian master/slave dialectic, but I am not interested in the relationship between the members in an oppositional set. Structuralist theorists and anthropologists have claimed and then disclaimed that societies organize themselves in binary sets, but more interesting to me than factual organization is how reality is perceived. Also, when an anthropologist or theorist says most concepts are defined by contrast with their opposites, I can't help but wonder whether we have quantifiable data along these lines.

I am hoping to find out whether these common-sense binary taxonomies correlate with cognitive/perceptive processes. Especially interesting to me is the question whether there are certain low-level cognitive strategies, though I am also interested in high-level processes (e.g. formal logic is mostly organized along binary lines, true/false). Since the cognitive sciences are not my field, I would be very grateful for references to both foundational studies and "raw" field work. General works on the reduction of reality to accommodate limited processing power are also welcome. Visual processes are not my primary interest, but if you can point to a single "best primer" on the filter that allows us to "see" an entire bookcase but not all book titles in a single glance, that would be a helpful bonus. I know nothing about the workings of biological neural networks, but if there is a binary selection at any point in their operation, I would be very grateful to see that explained in an accessible but peer-reviewed text. Finally, keywords to help me find more on this in scholarship search engines would be most helpful.

Answer 1

I think there is a misperception at work in your question. There is a wide variety of objects that we never perceive in such a binary manner: colors, fruit (apples, oranges, plums, ...), weather, and basically every other concrete objects. The only things we perceive in a binary fashion are abstract ideas! Good versus evil. Liberal versus conservative. And here we can ask about the reality (which is more complex) and the function of this simplification. The answer is: speed of decision making.

The basic binary classification is negation: x versus not x. The usefulness of this distinction becomes apparent when you put it in context: Is this berry edible or not? Do I want to trust this person or not? All binary opposites have a negation at their basis and divide the world into things you want to avoid and things you want to interact with.

And since human beings have to decide about one million things every day, they make life easier by using heuristics to make most of these decisions: simple rules based on experience, so you don't have to think before you vote for your next president.

The few binaries that are more or less objective, are categories of measurement: long/short, hot/cold, bright/dark, etc. That the physical reality underlying these measurements is maybe more complex (e.g. coldness depends not only on temperature, but also on the humidity of the air and the speed of the wind), is not relevant to the one-dimensionality of our sensation.

Night and day is not binary, since we have dawn and dusk and late afternoon and early morning and the dead of night as well to differentiate subjective time, relative visibility, quality of light, restedness/tiredness etc., that all make up the meaning of "night", "day" and other daytime names. The phrase "day and night" is binary only in certain phrases where they have symbolic meaning (e.g. "day" = good, "night" = evil, or "working night and day" = working all the time, without resting).

---

Studying the development of thinking styles in college students, Perry (1970) found that as they grow their thinking advances from a dualistic black-and-white perspective to a multiplicity of truths that allows for uncertainty. They also grow from a purely intellectual to an ethical perspective. In the course of this development, dualistic thinking can serve as an alternative to forward progression, and persons that already have developed to a higher level of thinking can use dualistic thinking as a "retreat" (Zang, 2002): "a student returns to a dualistic orientation to find security and the strength to cope with an overchallenging environment".

Binary thinking is related to personality disorders such as borderline personality and narcissism (Oshio, 2012): "The results indicated that thinking dichotomously may lead to wide-ranging personality disorders."

---

Sources:

• Oshio, A. (2012). An all‐or‐nothing thinking turns into darkness: Relations between dichotomous thinking and personality disorders. Japanese Psychological Research, 54(4), 424-429. doi:10.1111/j.1468-5884.2012.00515.x
• Perry, W. G. (1970). Forms of intellectual and ethical development in the college years: A scheme. New York: Holt, Rinehart and Winston.
• Zhang, L. F. (2002). Thinking styles and cognitive development. The Journal of Genetic Psychology, 163(2), 179-195.

Answer 2

Binary processes may be observed at various stages of intuitive and deliberative thought, which may in some cases plausibly be modelled as categories, but you’ll want to consider that on a case-by-case basis. Many of the relevant processes are binary simply because like formal logic, they involve toggle switches (true/false).

Heuristics

The simplifying processes alluded to in the question have been studied above all in the form of heuristics, i.e. mental strategies for judgement and decision-making that bypass a good deal of relevant computation in favour of simple cues that seem to (but may not) correlate well with success (Payne, Bettman, and Johnson 1993; Gigerenzer and Todd 1999; Czerlinski, Gigerenzer, and Goldstein 1999). A variety of models exist, most of which concern preferential decision-making, but categorization models also exist, and at any rate the two are closely intertwined. Most models involve binary processes in the selection process, particularly because judgement and decision-making involve matching objects against a criterion. Examples are the decision-making model Elimination by Aspects (Tversky 1972) and the categorization model Categorization by Elimination (Beretty, Todd, and Blythe 1997), which cycle through features in order of descending salience and/or effectiveness at narrowing down choice. Objects (options) that fail to meet the criterion under review are ruled out; it could thus be said that there is a binary matching principle at work even in cases where the cue (e.g. wine aroma) is not binary. Another popularly studied example is the recognition heuristic, which associates the known objects in a set with a higher criterion value than unknown objects in the same set: this strategy is modelled on a binary cue (match or no match, as opposed to a graded scale). The particular effectiveness of simple heuristics in situations with binary cues is discussed in Hogarth and Karelaia 2006.

It should be noted, however, that some of the most commonly mentioned studies of preferential decicion-making involve binary choice: which of the two? This binary feature is a research bias: binary choice is the minimal working example of preferential decision-making. Still, the recognition heuristic is of relevance to the question precisely because it does not provide a complete strategy for dealing with a variety of objects with different degrees of familiarity. It says simply that a first strategy can be to distinguish between completely unknown and not completely unknown, at which point the use of this heuristic is exhausted and the mind turns to other strategies. The real significance of this heuristic to the question therefore depends on its validity and documented application with non-binary sets and relative familiarity; the outcome of such tests would tell you a good deal about the real-world attestation of dichotomous thinking along these lines.

In the tradition of Payne and Gigerenzer, adaptiveness is a critical part of the heuristic process: environmental factors determine what reductive strategies we use. When simplifying reality through heuristic means, what features survive in our cognitive model of the world at the moment of judgement or decision-making will thus depend on why we are accessing these concepts in the first place. A binary set like day and night is more likely to occur to us when deciding whether to bring our bicycle lights when leaving the house, and less likely when scheduling a dentist appointment (though this too involves a binary set, available/unavailable).

Linguistics

The elimination models developed for decision-making have also been used to model memory access to semantic content. Rachel Giora has written a slew of articles advancing the idea that word senses that are salient in the mental lexicon are accessed first; other senses are activated if the first does not match the context (e.g. Giora 1999). The same matching paradigm could be extended across many aspects of syntax, for instance.

Keywords

That should be a start, but it’s a topic of some scope. Keywords could combine members of the following sets:

• contrast, contrastive, binary, dual, dualistic, dichotomous, dichotomy
• heuristic, cognitive, salience, simplify, simplification, judgment, judgement, decision

@Nick Stauner's comment and @what's answer contain references to significant additional avenues to investigate.

Bibliography

Berretty, Patricia M., Peter M. Todd, and Philip W. Blythe. 1997 “Categorization by Elimination: A Fast and Frugal Approach to Categorization.” Proceedings of the Nineteenth Annual Conference of the Cognitive Science Society. Eds M.G. Shafto and P. Langley. Mahwah, NJ: Lawrence Erlbaum Associates.

Czerlinski, Jean, Gerd Gigerenzer, and Daniel G. Goldstein. 1999. “How Good are Simple Heuristics?” Simple Heuristics That Make Us Smart. Eds. Gerd Gigerenzer and Peter M. Todd. Oxford: Oxford University Press. 97–118.

Gigerenzer, Gerd, and Peter M. Todd, eds. 1999. Simple Heuristics That Make Us Smart. Oxford: Oxford University Press.

Giora, Rachel. 1999. “On the Priority of Salient Meanings: Studies of Literal and Figurative Language.” Journal of Pragmatics 31: 99–129.

Hogarth, Robin M., and Natalia Karelaia. 2006. “ ‘Take-the-Best’ and Other Simple Strategies: Why and When They Work ‘Well’ With Binary Cues.” Theory and Decision 61: 205–49.

Payne, John W., James R. Bettman, and Eric J. Johnson. 1993. The Adaptive Decision Maker. Cambridge: Cambridge University press.

Tversky, Amos. 1972. “Elimination by Aspects: A Theory of Choice.” Psychological Review 79.4 (July): 281–99.