The Thinking Problem: Why Language Proficiency Is a Cognitive Asset
How the cognitive cost of non-native-language processing becomes a structural liability in strategic work
Many years ago, when I still worked in Capital Markets, I watched a vendor work through a strategic change in our operations. Fluent English. Articulate. Competent. But I noticed something else: he was managing two tasks simultaneously. Managing the strategy. And managing the language.
The room saw professional competence.
I observed precise orchestration of in-language thought and speech.
This is to reflect on the incorrect assumption that grounds how organizations structure global operations: if someone is fluent in English, they can think as effectively in English as they would in their native language.
The evidence says otherwise.
Bilingual cognition research is consistent. Fluency does not equal cognitive equivalence. When high-proficiency speakers operate in a non-native language under complexity and time pressure, measurable differences emerge. Not in intelligence. In cognitive architecture.
Keysar’s research on the foreign-language effect demonstrates this repeatedly. When bilinguals solve complex problems, negotiate under ambiguity, or make decisions with incomplete information in their non-native language, they show different patterns than in their native language. Bialystok’s work documents how language production in a non-native language occupies working memory otherwise available for reasoning. The more complex the task, the more pronounced the effect. Dewaele’s research on emotions and multilingualism adds another dimension: emotional regulation also operates differently across languages, affecting risk tolerance and judgment.
In that strategic planning session, the vendor was managing his thoughts in English while simultaneously managing the language. He could follow the discussion, articulate responses, perform at a level others would call fluent. And he was doing so at his full cognitive capacity.
The problem is not when competence is compromised by fluency.
Strategic reasoning under ambiguity and time pressure requires working memory availability, which non-native language processing consumes.
Ask anyone who is fluent in English—maybe even bilingual—and they will tell you how tired they get after a full day of meetings in a language that is not the one they use every day.
It is not because our thinking is degraded, but because the conditions for thinking have become more demanding.
This happens at every level of real-time cognitive demand. Multilateral discussions where a non-native speaker must track multiple voices, interpret subtext, formulate positions simultaneously. Negotiations where reasoning happens in real time as positions shift. Improvisational moments where prepared material ends and adaptive thinking begins. In each case, the cognitive tax is highest precisely when organizations depend most on full cognitive capacity.
Non-native-speaker leaders instinctively over-prepare. They understand the need and work through scenarios sometimes even in their native language, then retrieve and deliver in English. This converts real-time reasoning into retrieval, reducing cognitive load dramatically. But preparation has a clear boundary. It works as long as the discussion stays within the mapped territory. The moment negotiation shifts, an unexpected objection surfaces, or strategy requires improvisation, the cognitive cost reasserts itself. The leader working from preparation cannot simultaneously generate new thinking at full capacity.
There is a threshold where this changes: around B2 proficiency (CEFR), an upper-intermediate level where speakers can engage in extended, unscripted interaction. At this stage, they rely less on translation and begin to think more directly in the language. Working memory is not split between language production and content reasoning.
Organizations that recognize this invest in B2 proficiency for senior leaders not because it looks impressive on a CV, but because it is a structural variable affecting strategic performance. A leader who has crossed the B2 threshold participates differently in unscripted conversations: more willing to challenge, faster in adaptive reasoning, capable of the real-time thinking strategy requires.
Language proficiency in global organizations is not a soft skill. It is cognitive infrastructure.
When non-native-speaking leaders operate below B2, you absorb the cost of their reduced working memory in every strategic meeting, every negotiation, every decision made under uncertainty and time pressure. The calculation of what this costs is straightforward: count the difference between the strategy your leaders would generate in their native language and the strategy they actually generate under linguistic constraint. That difference is the organizational tax you have chosen to accept.
The deeper question is whether you have calculated it consciously or simply absorbed it as an invisible cost of global operations.
Before accepting the current model — language proficiency as communication skill — apply this to your most important strategic work. An executive team is meeting to decide on a capital allocation that will shape the next three years. Around the table: a mix of native and non-native English speakers, all high-performing, all certified proficient.
Now apply pressure. Challenge ideas in real time and observe. When you review the output, whose thinking was fully represented? Who was slightly behind — and never quite caught up? Have you calculated what it costs your organization to run strategic sessions where some participants are at 70% cognitive capacity because the remaining 30% is managing the language poorly?
Then this is not only a communication gap. It is a cognitive tax.
Investment in B2 proficiency is an investment in cognitive infrastructure, especially in the age of AI. Strategic reasoning must be fully available—in real time, under pressure, where it matters, including when detecting AI hallucinations.
The organizations that have calculated this are investing differently. Not in language training but as cognitive readiness and considering it a strategic asset.
Those who haven’t calculated it are already paying—silently forfeiting their best thinking in every meeting.
Works Cited
If you want to follow this line of thought further, here are the voices whose work helped shape the ideas I’ve shared here:
Keysar, Boaz, et al. “The Foreign-Language Effect: Thinking in a Foreign Tongue Reduces Decision Bias.” Psychological Science, vol. 23, no. 6, 2012, pp. 661-668.
Bialystok, Ellen, et al. “Bilingualism: Consequences for Mind and Brain.” Trends in Cognitive Sciences, vol. 16, no. 4, 2012, pp. 240-250.
Dewaele, Jean-Marc. Emotions in Multiple Languages. Palgrave Macmillan, 2010.
Content Authenticity Statement
AI tools were used selectively to support early-stage research and structural thinking. The strategic direction, original synthesis, interpretive judgment and final voice are entirely my own. This disclosure reflects a commitment to authorship clarity and emerging standards of transparency in content development.