We need to simplify the world in order to live in it. Every second, the brain receives a quantity of information it would be impossible to examine with equal depth. It classifies, compares, recognises patterns, eliminates possibilities and produces shortcuts. This capacity is fundamental. Faced with an immediate danger, a lengthy analysis of every nuance of the situation would even be counterproductive. Yet a mechanism efficient for deciding quickly becomes far less reliable when it serves to understand complex phenomena. This is where binary thinkingappears, also called dichotomous thinking: a tendency to organise reality around opposed categories — true or false, good or bad, success or failure, guilty or innocent, with us or against us.
Psychology has studied this pattern for several decades. In 2009, the psychologist Atsushi Oshio developed the Dichotomous Thinking Inventory, a tool designed to measure this tendency. His work distinguishes notably the preference for clear-cut categories, dichotomous beliefs and a way of evaluating certain situations along a logic of gain or loss. Binary thinking even has a certain contextual efficacy: it brings clarity, facilitates decision and makes a position immediately comprehensible. It offers a sharp structure to an ambiguous reality. Difficulties appear when that structure becomes an almost automatic grid applied to situations involving several causes, several levels of analysis and sometimes several simultaneous truths.
Two boxes give an impression of control
Uncertainty has a mental cost. An open situation obliges us to hold several hypotheses in mind, to accept that some information is missing, to revise our conclusions and to bear the idea that a definitive answer may remain out of reach for some time. Social psychology describes in particular the need for cognitive closure, that is, the desire to obtain quickly an answer stable enough to reduce ambiguity. The work of Arie Kruglanski and Donna Webster shows that this need can favour two movements: seizing rapidly on an available explanation, then clinging to it once it seems satisfactory. Its intensity varies with individuals and circumstances. Fatigue, pressure, conflict or urgency can reinforce the search for a clear answer.
This dynamic helps us understand the appeal of the binary without reducing it to a lack of intelligence. A simple answer soothes uncertainty. It makes it possible to know whom to support, whom to condemn, which cause to defend and which explanation to retain. It turns a situation that is hard to interpret into an immediately coherent narrative. Once that narrative is built, every new item of information must find a place within it. What confirms it is easily integrated. What weakens it demands more effort: modifying the narrative, admitting an error, or reintroducing an ambiguity the mind had already closed off.
Reality rarely works with such sharpness. A decision can produce benefits and damage at the same time. A person can be right on one point and badly wrong on another. An environmental policy can protect a species while displacing an ecological pressure elsewhere. A judicial decision can be legally coherent while remaining socially misunderstood. An innovation can solve a technical problem while creating a new dependency. Understanding these situations requires holding several levels of analysis at once.
We are not all equal in the face of uncertainty
The tendency to seek clear-cut answers varies considerably from one person to another. Personality, education, earlier experiences, cultural environment and immediate context influence how each person reacts to ambiguity. The same person can also prove very open in one domain and far more rigid in another, notably when their identity, their values, their security or their belonging to a group is directly concerned.
This variability is a reminder that binary thinking describes less a category of people than a mode of processing information liable to appear with different intensity depending on individuals and situations. Classing certain individuals among « binary people », as though they formed a homogeneous group, would in fact reproduce the very mechanism studied here: reducing a complex reality to too narrow a category.
When emotion narrows the field of the possible
Binary thinking is not solely a matter of the rational processing of information. Emotions likewise modify the way we perceive the available possibilities. Fear, insecurity, intense stress or certain traumatic experiences can strongly narrow the field of attention. When a person feels threatened, their organism naturally prioritises the rapid identification of danger and the responses that restore some form of security. In such moments, holding several contradictory interpretations becomes more difficult.
This dimension changes how we regard certain forms of rigidity. A person may cling to very clear-cut categories because they provide structure in an environment felt as unstable or dangerous. This effaces neither the errors of judgement nor their consequences, but it reminds us that cognitive rigidity can sometimes serve a protective function. Security facilitates access to nuance, because it leaves more mental room to examine several hypotheses, to defer a conclusion or to accept a zone of uncertainty.
This emotional dimension also avoids a frequent error: interpreting every simplification as intellectual laziness. Some rapid reactions correspond to adaptations forged in contexts where hesitation could be costly. They can become ill-suited when the context changes, while remaining deeply rooted in the way the person assesses danger.
When an idea becomes a camp
Binary thinking takes on another dimension when an opinion ceases to be merely an opinion and becomes a marker of belonging. Human beings build part of their identity through groups, values, communities and shared narratives. Some discussions then change in nature. A criticism addressed to an idea can be felt as a criticism of the group that carries it. Acknowledging the validity of an opposing argument can give the impression of ceding ground to the other camp. Correcting a belief sometimes becomes more costly socially than keeping an error.
Research on polarisation describes this slide toward a logic of « us against them ». A review published in Nature Reviews Psychology underlines the combined role of cognitive, motivational and social mechanisms in political polarisation. Individuals process information within an environment traversed by relationships, collective identities, relations of trust and representations of the adversary. All these elements influence how a piece of information is received, interpreted and possibly rejected.
It would nevertheless be too simple to explain every disagreement by the defence of a group identity alone. Prior beliefs, the trust granted to sources, personal experiences, the social context and the way information is presented also play a role. A person may reject a claim for several reasons at once. Reducing every disagreement to an identity bias would recreate precisely the mechanism we are trying to understand.
When disagreement becomes a judgement on the person
When a debate polarises, the arguments themselves can recede into the background. The interlocutor is sometimes classified before their reasoning has really been examined. A criticism of an environmental policy becomes proof of hostility toward ecology. A question about a court decision becomes a defence of the accused or an attack on the victim. A criticism of a technology becomes a rejection of progress. Conversely, defending an innovation may be enough to make someone the supposed representative of industrial interests they have never expressed.
This slide progressively eliminates intermediate positions. A person who refuses both camps becomes suspect in the eyes of each. Nuance can end up being interpreted as weakness, evasion or a strategy meant to mask one's true camp. Reasoning then turns in a closed circuit: the label assigned at the outset determines the interpretation of everything that follows.
One of the most revealing signs of binary thinking appears when a person attributes to their interlocutor the whole set of opinions associated with the group into which they have just placed them. A single position then suffices to reconstruct their entire system of thought artificially. The debate bears less on what they actually said than on what they are supposed to think. This mechanism saves an enormous amount of cognitive effort, at the price of a considerable loss of information.
When the binary becomes a strategy
Not all simplified oppositions arise spontaneously in the minds of those who spread them. The binary can also become a deliberate rhetorical tool. In political, activist, commercial or media communication, reducing a situation to two camps facilitates mobilisation: for or against, progress or regression, security or chaos, the people or the elites, defenders or enemies of a cause.
The strength of this strategy lies precisely in its simplicity. A narrative involving several causes, several responsibilities and several possible outcomes hardly mobilises anyone in a few words. A sharp opposition immediately produces an identity, an adversary and a direction. It turns a complex reality into a legible conflict, easy to transmit and emotionally powerful.
This distinction is essential. There is a binary thinking that is undergone, arising from cognitive, emotional or social shortcuts, and a constructed binarity, used to orient the perception of an audience. The two can then reinforce one another. Whoever manufactures a simplified opposition is not even obliged to believe in it entirely themselves. It is enough for them to know that it works.
In that case, simplification is no longer merely a limit of our way of thinking. It becomes a technique of framing. By defining the only two options supposedly available, whoever imposes the frame already influences part of the answer. Refusing that framing then requires going back upstream: checking whether the categories proposed really cover the whole problem or whether they were built to steer the choice.
Social networks offer particularly favourable ground
The internet created neither polarisation, nor tribalism, nor indignation. These phenomena existed long before the appearance of social networks. The platforms have, however, altered their speed, their scale and their visibility. A simple, emotional and immediately identifiable message often circulates more easily than a piece of reasoning that requires context. A few lines suffice to attach a piece of content to a camp. A complex analysis takes more time, while the attention economy favours rapid reactions.
A study published in 2024 in Communications Psychology, based on more than 6,000 observations among American users of Twitter, now X, observed that in the short term use of the platform was associated with an increase in outrage and polarisation, with different effects depending on usage. The authors also noted an increase in the sense of social belonging in certain forms of use. This combination deserves attention: the same environment can strengthen integration into a group and accentuate the felt distance from other groups.
Another phenomenon amplifies this dynamic. A study published in Nature Human Behaviour showed that users tend to overestimate the intensity of the moral outrage actually felt by the authors of the messages they read. This overestimation then feeds the impression that opposing groups are more hostile, more extreme and more aggressive than they really are. A society can thus begin to perceive itself as more divided than it is, and then adapt its behaviour to that perception.
Blaming the algorithms alone would nevertheless provide too comfortable an explanation. A large experiment conducted on Facebook during the 2020 American presidential election reduced by about a third the exposure of more than 23,000 participants to content from politically like-minded sources. The intervention did modify what users saw, notably by reducing certain uncivil content and certain sources that repeat false information, without producing a measurable change on eight indicators of political attitudes, including affective polarisation and ideological extremism. Platforms influence the informational environment, while human behaviours, identities, prior experiences and social structures continue to exert their own influence within it.
Polarisation and rigidity can reinforce one another
Another line of research offers an interesting light: the relation between polarisation and cognitive flexibility. Researchers have proposed that a strongly polarised environment favours greater rigidity in the way information is processed, while that rigidity can itself sustain polarisation. Conflict, conformity to the group and the simplification of information can form a reinforcing loop. In the short term, strong internal coherence facilitates a group's coordination. In the longer term, it reduces its capacity to integrate information coming from outside.
This logic extends well beyond politics. It can appear in a family conflict, a company, a scientific community, an activist association, a discussion about health, the environment or justice. The more a subject becomes identity-laden or emotional, the more the psychological cost of changing position may rise. A collective sometimes ends up valuing coherence with its own narrative more highly than the capacity to correct it.
The phenomenon takes on another scale when it settles into the institutions themselves. A complex society depends on systems capable of examining contradictory information: science, justice, journalism, medicine, public expertise. Their reliability rests largely on the possibility of revising a conclusion when new data appear. A culture that turns every correction into an admission of weakness encourages immobility and ends up making correction more costly than error.
Changing one's mind also has a social cost
Valuing the revision of a position intellectually remains insufficient when the environment penalises it. In political, media or professional life, an old statement can be brought back years later to accuse someone of inconsistency, even when their change of position rests on new information. Social networks preserve stated positions, confront them, and often reward displayed constancy more than the evolution of reasoning.
Organisations can reinforce the same rigidity. When admitting an error threatens a reputation, a career, a hierarchical position or the image of an institution, everyone receives an additional incentive to defend their earlier decisions. An environment able to distinguish error, fault, uncertainty and reasoned revision offers, by contrast, more room to correct a trajectory.
Cognitive flexibility thus depends as much on individual competences as on the social conditions allowing them to be exercised. Asking individuals to acknowledge their errors more readily without modifying the environments that punish them for doing so amounts to ignoring part of the problem.
Nuance effaces neither facts nor responsibilities
Leaving the binary behind in no way obliges us to place all positions on the same level. Some claims are false. Some behaviours are dangerous. Some evidence is far more solid than other evidence. Some moral limits require a clear response. Complex thinking adds an essential distinction between passing judgement on a precise element and reducing a whole reality to that judgement.
One can condemn an act while seeking the mechanisms that made it possible. Acknowledging the benefits of a technology remains compatible with studying its negative consequences. Defending the protection of the living permits criticism of an ineffective environmental measure. Acknowledging individual responsibility can coexist with an analysis of the social context. These propositions become compatible as soon as several levels of causality are examined simultaneously.
Nuance sometimes demands more intellectual firmness than the binary. It requires resisting the immediate satisfaction offered by a simple conclusion. It obliges us to distinguish what is established, probable, possible and unknown. It also demands revising one's own reasoning when the available elements change, without turning that revision into a personal defeat.
Increasing the resolution of our thinking
A photograph made of two colours can show a silhouette. To distinguish textures, distances and details, the resolution must be increased. Reasoning works in a comparable way. Between true and false may appear the plausible, the incomplete, the contextual, the uncertain or the partly exact. Between responsible and innocent lie different levels of responsibility. Between success and failure may appear contradictory results. Between two camps there are often thousands of people who share certain findings while diverging on their causes, their priorities or their solutions.
Leaving binary thinking behind consists in multiplying hypotheses before narrowing the field of the possible. It requires separating a person from one of their ideas, a result from the intention that produced it, a correlation from a causation, a criticism from a wholesale rejection. It also requires recognising the situations in which two apparently opposed claims in fact describe two different parts of the same system.
Human thought will always need categories. Removing all simplification would be unrealistic and inefficient. What is at stake is knowing when to simplify and when to reopen the model. An urgent decision may require two options. Understanding a social, ecological or human system often calls for more resolution.
Ways of resisting binary thinking
Understanding the mechanism is already a first protection, but certain habits make it possible to keep reasoning open for longer. Tolerating uncertainty amounts to accepting that a conclusion may remain provisional and that the absence of an immediate answer does not mean that reflection has failed. Verbalising presuppositions makes it possible to check what we are really attributing to the other person before responding to a position they have never defended. Valuing reasoned revision consists in recognising that a change of mind based on new information may attest to greater rigour than the obstinate defence of an old position.
Another method consists in deliberately seeking a third possibility whenever a situation seems to reduce spontaneously to two options. This third hypothesis may be imperfect, provisional or ultimately rejected. Its value lies in the displacement it produces: the first two possibilities cease to appear as the only imaginable forms of the real.
This method nevertheless has an important limit. Seeking a third possibility never means that the truth necessarily lies between two opposed positions. That would be to fall into the fallacy of the golden mean, which consists in holding that an intermediate position would necessarily be more reasonable simply because it lies between two extremes. Some claims are better supported than others, sometimes by a very wide margin. A hypothesis gains no value from its position between two camps.
The « third box » must remain a tool of exploration, never a rule of conclusion. It serves to check whether the proposed alternative really covers the whole range of possibilities. Once the field is reopened, each hypothesis must be confronted with the same requirements of evidence, coherence and confrontation with the real. Sometimes the third way will provide a better explanation. Sometimes one of the two initial proposals will remain clearly superior.
Finally, distinguishing levels of analysis avoids many false oppositions. Two people may seem to contradict one another while speaking about different realities. One describes an individual responsibility, the other a collective mechanism. One observes an immediate consequence, the other a structural effect. One speaks of intention, the other of outcome. Specifying the level at which each claim applies sometimes reveals that they can be valid at the same time.
Complex thinking promises no comfortable answer
Resisting binary thinking does not consist in systematically producing more nuance, in seeking a compromise between all positions or in artificially keeping every hypothesis open. Some situations demand a decision. Some data clearly eliminate certain possibilities. Some decisions must be taken despite uncertainty.
Complex thinking consists above all in avoiding closing off reasoning before the available elements permit it. It accepts that a firm conclusion may emerge after an open analysis. It distinguishes the necessity of deciding from the temptation to simplify. It also recognises that the same person can become very binary in some contexts while retaining great suppleness in others.
We live in an age in which anyone can produce an opinion in a few seconds, broadcast it immediately and receive almost at once the approval of those who share it. Institutions, groups, emotions and communication strategies can reinforce this dynamic. Resistance to binary thinking then rests neither on a moral quality reserved for a few nor on simple individual will. It calls for cognitive competences, an environment that permits revision, a certain emotional security and a vigilance toward the frames imposed on us.
In this environment, one discreet intellectual capacity becomes particularly precious: holding several possibilities in mind long enough to avoid choosing the wrong one too early.
The opposite of binary thinking is neither indecision nor an absence of convictions. It is the capacity to hold convictions without forcing the real into two boxes.
Principal sources
• Atsushi Oshio, Development and validation of the Dichotomous Thinking Inventory, 2009.
• Arie W. Kruglanski and Donna M. Webster, work on the need for cognitive closure and the mechanisms of seizing et freezing.
• John T. Jost, Delia S. Baldassarri and James N. Druckman, Cognitive–motivational mechanisms of political polarization in social-communicative contexts, Nature Reviews Psychology, 2022.
• William J. Brady et al., Overperception of moral outrage in online social networks inflates beliefs about intergroup hostility, Nature Human Behaviour, 2023.
• Brendan Nyhan et al., Like-minded sources on Facebook are prevalent but not polarizing, Nature, 2023.
• Victoria Oldemburgo de Mello, Felix Cheung and Michael Inzlicht, Twitter (X) use predicts substantial changes in well-being, polarization, sense of belonging, and outrage, Communications Psychology, 2024.