The Mental Health Cost of Code-Switching

Before the meeting begins, you hear your voice change.

You slow down. You reconsider a phrase that would have come naturally with family or friends. You check whether your expression looks interested but not irritated, confident but not intimidating. You think about how much emotion is safe to show. You may have already considered your hair, your clothes, whether coworkers will pronounce your name correctly, and whether correcting them is worth what could follow.

None of these decisions necessarily appears dramatic from the outside. Some happen so quickly that they barely register as decisions. Yet they require attention.

For many Black Americans, code-switching is part of moving through workplaces, schools, healthcare settings, stores, neighborhoods, and other spaces where being understood is only one concern. The other questions may be harder: Will I be believed? Will I be treated as competent? Will directness be called aggression? Will comfort be mistaken for carelessness? Will I be evaluated on what I said, or on whether I made other people comfortable while saying it?

Code-switching can be useful. It can protect access, help someone communicate across groups, and make a difficult environment easier to move through. It can also become exhausting when acceptance seems to depend on continually calculating which parts of yourself are safe to show.

Research directly connecting racial code-switching to mental-health outcomes is still limited. Much of the available work is correlational or examines perceived professionalism, authenticity, emotional labor, burnout, and related psychosocial experiences rather than diagnosed conditions. The broader evidence on racism and vigilance helps explain the context, but it does not prove that code-switching itself causes illness.

What code-switching means

The term code-switching originally described moving between languages or language varieties within a conversation or across settings. In everyday use, it has expanded to include changes in speech, tone, expression, appearance, behavior, and emotional presentation in response to a social environment.

For Black Americans, this may include shifting between African American English and standardized American English, changing cadence or vocabulary, making a name sound more familiar to White colleagues, choosing a different hairstyle, limiting cultural references, softening disagreement, or monitoring facial expressions and body language.

African American English is a systematic language variety with its own grammatical patterns and history, rather than careless or broken English. Treating one language variety as naturally intelligent or professional while treating another as deficient is a social judgment, not a linguistic fact. In a commentary drawing together prior theory and research, Darin Johnson and colleagues propose that linguistic racism can make code-switching a response to stereotype threat and add to the mental work involved in deciding how to speak (Johnson et al., 2022).

The broader meaning of code-switching can become so loose that it describes almost any change in behavior. That is not helpful. Speaking differently with a child than with a supervisor is ordinary social adjustment. Racial code-switching becomes a more specific issue when someone modifies identity-related behavior because dominant norms carry unequal consequences.

“Doesn’t everyone do this?”

In one sense, yes. Most people adjust how they communicate depending on the audience. We do not speak to a judge, a close friend, and a toddler in exactly the same way.

The difference is not simply whether adaptation occurs. The difference is what is at stake, how much of the self must be monitored, and who is expected to do most of the adapting.

A White employee may become more formal during a presentation. A Black employee may become more formal while also calculating whether their natural speech will be heard as less educated, whether firmness will activate an “angry” stereotype, or whether a cultural hairstyle will be treated as less professional. Both people are responding to context. They are not necessarily carrying the same social risk.

Ordinary flexibility can be energizing or neutral. Repeated self-protection may require a different level of vigilance.

Code-switching can affect professional judgments

It would be inaccurate to say code-switching is useless. Research suggests that it can affect the professional judgments people are trying to navigate.

Courtney McCluney and colleagues conducted experiments in which Black and White participants evaluated fictitious Black employees. The scenarios bundled several cues: how an employee spoke, the employee’s explanation of a name choice, and descriptions of hairstyle and speech style. In these controlled scenarios, code-switching generally increased perceived professionalism. Support was stronger for changing speech than for changing a name. The hairstyle effect differed by participant race among women, and Black participants rated the non-code-switching employee as more professional than White participants did (McCluney et al., 2021).

Those were controlled scenarios about perceived professionalism, not records of hiring, promotion, or salary decisions. Still, the findings reveal why advice to “just be yourself” can feel detached from reality. People adapt because they have learned that presentation affects how others interpret the same competence.

There are legitimate workplace expectations. Colleagues need to communicate clearly, treat one another respectfully, and meet the requirements of their roles. The problem begins when “professional” quietly means culturally familiar to the dominant group. Clarity is job-relevant. Making a name sound less Black is not. Respect is job-relevant. Natural hair is not evidence of disrespect. Giving useful feedback is different from asking someone to make their identity less visible.

The calculation itself takes energy

Code-switching involves more than performing the adjusted behavior. It can require reading the room before deciding what to do.

You may monitor who is present, how much authority they hold, what happened the last time someone spoke plainly, and whether this is a setting where humor, frustration, disagreement, or cultural familiarity will be understood. You may rehearse a sentence, edit it while speaking, and review it again afterward.

Johnson and colleagues studied self-reported code-switching in two online samples of Black Americans, with 342 participants in one study and 411 in the other. Stronger stereotype threat and need to belong were associated with more self-reported code-switching, while greater social connectedness was associated with less. Both assimilationist and nationalist racial ideologies were associated with more switching. Black identity, but not American identity, was associated with switching, while mentalizing tendencies were not. Because these were correlational surveys, they cannot tell us that any one factor caused code-switching. They show that these communication decisions are connected to identity and relationships, not merely vocabulary (Johnson et al., 2026).

That mental work can continue after the interaction. Was I too quiet? Too direct? Did I sound angry? Did I let that comment pass because I was being strategic, or because I froze? Should I correct what they said, or will that become the only thing anyone remembers about the meeting?

An individual exchange may take only a few minutes. The burden comes from repetition.

Vigilance has a cost

Code-switching can overlap conceptually with a broader process researchers call vigilance: preparing for possible mistreatment, watching what you say, monitoring appearance, avoiding certain situations, and anticipating how prejudice may show up. The vigilance study below did not measure code-switching, so it offers context rather than a direct test of its effects.

Hedwig Lee and Margaret Hicken examined vigilance using a population-representative sample of 3,105 adults in Chicago. Among Black adults, greater vigilance was associated with more depressive symptoms, poorer self-rated health, and greater risk of chronic health conditions. The study was cross-sectional, so it cannot establish that vigilance caused those outcomes. It also used data collected between 2001 and 2003 in one city. Its value lies in showing that protective preparation and health can be related even when no single discriminatory event explains the entire burden (Lee & Hicken, 2016).

The larger racism-and-health literature supports taking the surrounding context seriously. A systematic review and meta-analysis by Yin Paradies and colleagues synthesized 293 studies reported in 333 articles, most from the United States. Self-reported racism was associated with poorer mental and physical health, with stronger relationships for mental-health outcomes. Associations were stronger in cross-sectional than longitudinal studies and in non-representative than representative samples, reasons to be cautious about causal interpretation (Paradies et al., 2015). This evidence concerns racism broadly, not code-switching specifically.

The mental-health burden does not come from Black culture or from the ability to communicate across cultures. It can come from having to remain alert to environments in which misunderstanding, stereotyping, or discrimination carries real consequences.

The question of authenticity is complicated

People often describe code-switching as feeling like they are hiding who they are. That experience is real, but it is not universal.

Nirupika Sharma and Serena Chen studied racial and ethnic code-switching in a cross-sectional pilot and a 14-day diary study. The diary included 233 people of color in the United States, half of whom identified as Black or African American. Participants reported code-switching on an average of 9.45 of the fourteen days. On days when they reported more code-switching than usual, they also reported lower subjective authenticity and less shared reality with their racial or ethnic in-group, along with greater shared reality with White out-group members. The between-person pattern was different: people who habitually switched more reported greater shared reality with both groups. Among people lower in in-group identification, greater habitual switching was also associated with greater subjective authenticity. Exploratory analyses did not find that one day’s switching predicted the next day’s outcomes. These relationships varied with in-group identification, and the study cannot establish causation (Sharma & Chen, 2026).

A person may feel fully authentic while choosing different language for different settings. Another person may feel increasingly divided between the version of themselves that succeeds publicly and the version that feels known privately. Someone else may experience both: pride in being able to move between worlds and anger that one world demands so much adjustment.

Authenticity should not become another standard Black people are required to meet. “Bring your whole self to work” sounds inviting, but disclosure is not always safe, and nobody owes an employer unrestricted access to their identity. Choosing privacy or strategy does not make a person dishonest.

The more useful question is whether the choice feels available. Can you decide when to adapt, or does every interaction feel like a test you cannot afford to fail?

Black women may be managing more than one stereotype

Gender can shape which stereotypes a person is trying to anticipate.

Black women may be expected to appear confident without seeming intimidating, warm without being dismissed, strong without needing support, and composed even when responding to disrespect. A direct tone may be interpreted through the stereotype of the “angry Black woman.” Showing strain may conflict with expectations that Black women remain endlessly capable.

A 2025 study of 289 Black women examined the strong Black woman stereotype, racial and gender identity, mental-health symptoms, and identity shifting in academic and professional settings. Identity shifting is a related but broader construct than racial code-switching. Greater endorsement of the strong Black woman stereotype was associated with greater use of identity-shifting strategies. Gendered racial identity centrality was not significantly related to shifting, and anxiety and depression did not moderate the association between the stereotype and shifting. The study was cross-sectional and recruited participants through email, flyers, texts, and social media, so it does not establish direction or represent every Black woman (Hall et al., 2025).

Black women do not respond in one predictable way, but racial and gender expectations can operate together. Advice that ignores this may tell someone to be more assertive without acknowledging how differently that assertiveness may be received.

It can affect connection on both sides

Code-switching may help someone enter a space while making it harder to feel at home there.

Around White colleagues, a person may wonder whether acceptance depends on the edited version of them. Around Black friends, relatives, or coworkers, they may worry about seeming distant, overly formal, or disconnected from the community. That outcome is not universal: Sharma and Chen found that people who habitually switched more reported greater shared reality with their in-group, even though the day-to-day association ran in the opposite direction. The person can still end up carrying other people’s assumptions from both directions.

That does not mean Black communities uniformly reject code-switching. Black families have long taught children how to move safely and successfully through institutions that may judge them unfairly. Those lessons can be acts of care. They can also carry grief because the lesson exists for a reason.

Recent research on online interracial contact illustrates another side of the burden. Across three studies involving 1,107 participants in majority-White countries, the researchers studied Black participants in all three studies and included White participants for comparison only in the second. In that study, Black participants reported more code-switching, emotional labor, burnout from interracial contact, and desire for anonymity than White participants. Across the research, stigma consciousness and perceived discrimination helped explain part of the relationship between these experiences and anonymity. The studies relied on self-report and hypothetical or online contexts, so they should not be generalized to every interracial relationship (Nitschinsk et al., 2026).

Code-switching is not a diagnosis

The DSM-5 does not identify code-switching or racial stress as a mental disorder. It does include Target of (Perceived) Adverse Discrimination or Persecution as a condition that may be a focus of clinical attention. That category recognizes that perceived or experienced discrimination can matter in treatment without turning the person’s response into pathology.

Some experiences around code-switching can resemble anxiety. A person may rehearse, avoid scrutiny, monitor how they appear, or worry about negative evaluation. Social anxiety disorder requires persistent fear or avoidance that causes significant distress or impairment and is out of proportion to the actual threat and sociocultural context. Caution that matches what someone has actually seen penalized is not, by itself, evidence of a disorder. Accurate awareness of bias and an anxiety disorder can also coexist; assessment should examine both rather than assuming one rules out the other.

Racism can also be experienced as traumatic, but posttraumatic stress disorder has specific exposure and symptom requirements. DSM-5 Criterion A requires exposure to actual or threatened death, serious injury, or sexual violence. Many discriminatory experiences do not meet that threshold, and an experience does not have to qualify for PTSD to be painful, destabilizing, or worthy of care.

A culturally responsive assessment asks what happened, what the person reasonably expects could happen, how much distress or impairment is present, and how culture and discrimination shape the situation. The DSM-5 Cultural Formulation Interview offers a structured way to ask about cultural identity, stressors, supports, and barriers to care. Assessment should not assume that vigilance is irrational simply because the clinician has not faced the same risk.

When the cost is becoming too high

There is no universal amount of code-switching that becomes unhealthy. The impact depends on the setting, the degree of choice, the consequences, and the person.

It may be worth paying closer attention when:

• you feel depleted after routine interactions because you were monitoring yourself throughout them;
• you replay conversations for hours, looking for the moment you may have confirmed a stereotype;
• you avoid opportunities because the performance required to enter the room feels unbearable;
• you feel that success belongs to a version of you that nobody close to you recognizes;
• anger, sadness, headaches, muscle tension, disrupted sleep, or dread repeatedly follow certain environments;
• you no longer have places where you can speak, dress, disagree, laugh, or rest without calculating how you will be interpreted;
• workplace or school expectations have become impossible to separate from racialized treatment; or
• the strain is affecting relationships, concentration, health, or your willingness to seek help.

These signs do not prove a diagnosis. They may indicate that adaptation is taking something from you. Recurrent headaches, sleep disruption, muscle tension, or other physical symptoms may also deserve medical assessment rather than being attributed to stress without evaluation.

Protecting yourself without blaming yourself

There is no honest mental-health advice that can promise to remove racism through better coping. The responsibility for biased environments belongs to the institutions and people maintaining them.

Still, people deserve support while living in the world as it is.

It can help to become more specific about the pattern. Notice where you change, what you change, what you believe the adjustment protects, and how you feel afterward. This is a way to distinguish choices that serve you from demands that steadily drain you, not an exercise in catching yourself doing something wrong.

Preserve relationships and spaces where explanation is unnecessary. Time with people who understand your language, humor, references, frustration, and joy can reduce the feeling that every interaction requires translation.

Allow strategy without shame. You may decide that a particular meeting, interview, classroom, or medical appointment is not the place to challenge every expectation. Protecting your livelihood or safety is not consent to the bias that made the calculation necessary.

Also allow anger to carry information. Anger may be telling you that an expectation is unfair, a boundary was crossed, or you are tired of making yourself easier to receive. It does not have to control the next decision to be valid.

Recovery should fit the actual demand. After a day of heavy monitoring, you may need solitude, movement, music, prayer, laughter, conversation with someone who understands, or simply time in which nobody is evaluating you. Rest is not a complete answer to discrimination, but it can interrupt the expectation that you remain composed forever.

Workplaces have responsibilities too

Organizations often celebrate authenticity while leaving narrow definitions of professionalism untouched.

A better approach examines which standards are necessary for the work and which merely reflect familiarity with White, middle-class cultural norms. Feedback should identify the behavior, its job-related impact, and the expected result. “Your analysis needs a clearer recommendation” is actionable. “You need more polish” may conceal an unexamined cultural preference.

Managers can learn to pronounce names, evaluate outcomes consistently, question whether hairstyles or language patterns are being treated as competence signals, and notice who receives labels such as aggressive, articulate, intimidating, or surprisingly professional. They should not depend on Black employees to provide unpaid education or disclose personal experiences to prove that a problem exists.

People should remain free to adapt. Workplaces are responsible for making that adaptation less compulsory and ensuring that difference carries fewer penalties.

When therapy may help

Therapy cannot make an unfair environment fair. It can offer a place to put down the performance, name what has happened, and examine what constant vigilance is doing to your mood, sleep, body, relationships, and sense of self.

Culturally responsive therapy should not train a Black client to tolerate more racism or treat every concern about bias as distorted thinking. It can help separate present danger from old expectations when that distinction is useful, strengthen choices about where to spend emotional energy, address anxiety or depression when present, and support decisions about work, relationships, boundaries, and recovery.

At Coalesce Family Services, individual counseling offers a confidential space to discuss racial stress, workplace strain, anxiety, identity, and the pressure to remain acceptable in environments that may not fully accept you. Individual counseling is available through telehealth across South Carolina and Virginia.

Research note: Code-switching is defined differently across linguistics, psychology, organizational research, and everyday conversation. Several studies cited here are cross-sectional, use self-report, or examine controlled scenarios rather than real employment decisions. Associations with stress, authenticity, or mental health do not prove that code-switching caused those outcomes. Experiences also differ across Black communities and by gender, age, class, region, disability, sexuality, profession, and the degree of choice involved. This article therefore treats code-switching as a context-dependent strategy shaped by unequal environments, not as a disorder or a universally harmful behavior.