When Experience Becomes Evidence

Confirmation bias, reverse inference, and the stories we tell about psychedelics, meditation, and AI

Science is often imagined as a disciplined escape from belief. In practice, it is more fragile than that. Scientists are human beings before they are methodologists. They have experiences, communities, intellectual lineages, incentives, and hopes. Sometimes these forces inspire great science. At other times, they quietly bend the arc of inquiry towards conclusions that were desired before the evidence arrived. This is not fraud. It is often more ordinary and more dangerous: confirmation bias dressed as discovery.

Confirmation bias refers to the tendency to seek, interpret, and remember evidence in ways that support what we already believe. Peter Wason’s classic work on hypothesis testing showed how readily people fail to test alternatives that might disconfirm their favoured hypothesis [1]. Raymond Nickerson later described confirmation bias as a “ubiquitous phenomenon,” not an occasional defect of poor reasoning but a general vulnerability of human cognition [2]. The implication for science is sobering. The scientific method is not powerful because scientists are naturally objective. It is powerful because it creates structures that make self-deception harder.

Richard Feynman put this with characteristic clarity in his 1974 lecture on “Cargo Cult Science”: “The first principle is that you must not fool yourself, and you are the easiest person to fool” [3]. That warning is especially relevant in domains where the scientist’s personal experience is entangled with the object of study. Psychedelic science, meditation research, and contemporary artificial intelligence all provide timely examples. Each field contains genuine promise. Each has produced important insights. Yet each also risks being shaped by communities of believers who are, understandably, tempted to convert transformative personal or professional experience into universal prescription.

From Experience to Evidence: The Psychedelic Temptation

Consider psychedelic science. Many researchers and advocates came to the field because psychedelics changed their lives, or because they witnessed striking therapeutic transformations. That motivation is not illegitimate. Indeed, some of the most important scientific programmes begin with astonishment. The modern revival of psilocybin research was catalysed by carefully conducted studies showing that psychedelic experiences can carry substantial personal meaning and, in some clinical contexts, therapeutic potential [4]. But precisely because these experiences can be profound, they create an epistemic hazard. The person who has benefited from psychedelics may not simply ask, “What do these compounds do, under which conditions, for whom, with what risks, and by what mechanisms?” They may instead begin from a tacit conviction: “These experiences reveal something deeply healing, and the science must now show why.”

That shift is subtle but consequential. When prior belief becomes the gravitational centre of research, ambiguous findings are easily recruited as support. A change in brain connectivity becomes “increased integration.” A subjective report becomes “ego dissolution.” A therapeutic response becomes evidence of “neural reset.” These interpretations may sometimes be right, but they are not automatically licensed by the data. The danger is reverse inference: moving too quickly from an observed neural or behavioural marker to a richly specified psychological explanation. Russell Poldrack’s influential discussion of reverse inference in neuroimaging made this point carefully: such inferences are not impossible, but they require additional evidence, stronger constraints, and awareness that most neural measures do not map uniquely onto a single mental process [5]. The same caution applies beyond neuroimaging. A reported mystical experience does not, by itself, establish therapeutic mechanism. A reduction in symptoms does not, by itself, prove the metaphysical or psychological story attached to it.

Mindfulness, Meaning, and the Problem of Beautiful Priors

Meditation science faces a related problem. Many leading scientists in this area are themselves meditators, and many entered the field because contemplative practice had value in their own lives. Again, this can be a strength. Lived familiarity can sharpen questions and prevent superficial measurement. But it can also narrow scepticism. The claim that meditation reduces stress is one thing. The claim that it is a broadly applicable route to compassion, insight, neural health, social harmony, or human flourishing is another. The evidential burden increases as the claims expand. Yet popular and scientific discourse often moves too quickly from modest effects to sweeping narratives.

Critical reviews of mindfulness and meditation research have pointed to definitional problems, variable intervention quality, expectancy effects, inadequate controls, publication bias, and under-reporting of adverse experiences [6]. These concerns do not mean meditation is ineffective. They mean the science must resist being absorbed into the promotional language of the practice community. A meditator-scientist has a special responsibility to ask not only, “How did this help me?” but also, “Who does it not help? When can it harm? Which effects are specific to meditation rather than expectation, community, teacher contact, relaxation, or demand characteristics? Which claims survive rigorous controls?”

The Neural Network Gospel

The same structure appears in artificial intelligence, although in a different form. The spectacular success of deep learning has produced a generation of researchers and technologists who have witnessed neural networks solve problems once thought to require explicit rules, symbolic structures, or handcrafted representations. This success is real. It has transformed computer vision, language modelling, protein prediction, reinforcement learning, and scientific discovery. But success can harden into ideology. Some deep learning advocates now speak as if scale, data, and gradient descent are not merely powerful tools but the only serious path to intelligence. In this rhetoric, alternative approaches are treated less as hypotheses to be tested than as historical residues to be swept away.

Here too, the problem is not enthusiasm. The problem is premature exclusivity. The history of science is full of cases where a successful method becomes an imperial metaphor. Because deep networks work remarkably well in some domains, it becomes tempting to infer that biological cognition must be deep learning, that intelligence is prediction at scale, that embodiment is optional, that symbolic reasoning is obsolete, or that interpretability is a secondary concern. These conclusions may turn out to be partly right, but they do not follow automatically from benchmark success. A method can be powerful without being complete. A model can predict without understanding. A system can generate fluent behaviour without sharing the causal structure of human cognition.

Richard Sutton’s “Bitter Lesson” is often invoked to argue that general methods that scale with computation ultimately outperform approaches built around human knowledge [7]. It is an important lesson, but not the only one. A lesson about the power of scale is not automatically a theory of intelligence. Nor does it remove the need to understand representation, embodiment, agency, causality, social learning, biological constraint, or meaning. Deep learning’s success should expand scientific imagination, not shrink it into a single prescription.

The Loop: Belief, Data, Story, Repeat

The irony is that each of these fields often defines itself against dogma. Psychedelic science casts itself as a liberation from narrow psychiatry. Meditation science casts itself as a bridge between contemplative wisdom and empirical psychology. Deep learning casts itself as a triumph over brittle hand-engineering. Yet all three can reproduce the very tendency they oppose: the conversion of a partial truth into a total worldview.

This is where confirmation bias and reverse inference reinforce each other. Confirmation bias selects the evidence we find compelling. Reverse inference supplies the story that makes the evidence feel explanatory. Together, they create a loop. We begin with a belief: psychedelics heal by dissolving the ego; meditation trains attention and compassion; neural networks reveal the general principle of intelligence. We then observe data: altered connectivity, changed questionnaire scores, improved task performance, larger models producing better outputs. Finally, we infer that the data confirm the belief. The circle closes, and the hypothesis becomes harder to question because it now appears “evidence-based.”

But evidence-based science is not the same as evidence-decorated belief. The difference lies in whether evidence has the power to surprise us, constrain us, and force revision. A scientific claim should not merely accumulate supportive examples. It should expose itself to meaningful risk. What would count against the claim? What alternative explanations remain viable? What would make us reduce our confidence? Which analyses were specified in advance? Which outcomes were primary? Which negative findings were published? Which measures were chosen because they were theoretically diagnostic, and which because they were convenient or likely to produce a story?

Evidence-Based or Evidence-Decorated?

John Ioannidis argued that the probability a research finding is true depends not only on statistical significance, but also on bias, power, multiplicity, and the ratio of true to false hypotheses being tested [8]. This point is particularly important in fast-moving fields with high public interest. When a field is exciting, underpowered positive findings travel quickly. Negative findings remain unpublished. Ambiguous findings become narratives. Small mechanistic studies are used to support large translational claims. Investors, journalists, universities, and advocacy groups amplify the most attractive version of the story. The scientific ecosystem itself becomes a confirmation machine.

What, then, is the remedy? It is not cynicism. Cynicism is just confirmation bias with a negative prior. Nor is the remedy to exclude believers from science. Many transformative fields are built by people who care deeply about the phenomenon under study. The remedy is disciplined pluralism: a culture in which enthusiasm is permitted but not allowed to substitute for inference.

What Better Science Would Look Like

For psychedelic science, this means separating therapeutic efficacy from mechanism, and mechanism from metaphysics. It means studying adverse events, expectancy, therapist effects, set and setting, participant selection, and long-term outcomes with the same seriousness as acute mystical experience. It means asking whether impressive subjective experiences are necessary, sufficient, neither, or merely correlated with benefit. It means resisting the temptation to treat every neural change as a signature of healing.

For meditation science, it means defining practices precisely, using credible active controls, measuring harms, distinguishing state effects from trait effects, and avoiding the assumption that ancient practices must map neatly onto modern constructs. It means allowing the possibility that meditation is useful, but not universally useful; powerful, but not always benign; psychologically meaningful, but not automatically mechanistically understood.

For artificial intelligence, it means distinguishing engineering success from scientific explanation. Scaling laws are important, but they are not a theory of mind. Predictive performance is important, but it is not a complete account of intelligence. Neural networks are extraordinary tools, but their success should invite deeper inquiry, not methodological monopoly. The right question is not whether deep learning works. It plainly does. The right question is what kind of understanding its success gives us, what it leaves unexplained, and where alternative principles remain necessary.

Let the Evidence Surprise Us

Across all these domains, the deeper issue is intellectual humility. Scientists must be especially suspicious of findings that flatter their priors, vindicate their identities, or reward their communities. The most dangerous result is not the one that contradicts us. It is the one that confirms us too easily.

Good science is not self-serving. It is not the ritual production of evidence for conclusions we already hoped were true. It is a method for being answerable to the world. That requires more than data. It requires adversarial collaboration, preregistration where appropriate, open materials, publication of null results, serious engagement with critics, and conceptual restraint. It requires asking whether our favourite explanation is uniquely supported by the evidence, or merely compatible with it.

The goal is not to remove human passion from science. Without passion, many difficult fields would never advance. The goal is to ensure that passion remains upstream of the question, not downstream of the answer. Personal transformation can motivate research, but it cannot validate a theory. Technological success can inspire a programme, but it cannot license universal prescription. A beautiful hypothesis can guide inquiry, but it must still be vulnerable to defeat.

The test of a scientific culture is not how confidently it celebrates its successes. It is how willingly it investigates its own temptations. Psychedelics may transform some lives. Meditation may cultivate important capacities. Neural networks may reveal powerful principles of learning. But none of these possibilities frees us from the first principle. We must not fool ourselves. And we are still the easiest people to fool.

References

[1] Wason, P. C. (1960). On the failure to eliminate hypotheses in a conceptual task. Quarterly Journal of Experimental Psychology, 12(3), 129–140.

[2] Nickerson, R. S. (1998). Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2(2), 175–220.

[3] Feynman, R. P. (1974). Cargo Cult Science. Caltech commencement address.

[4] Griffiths, R. R., Richards, W. A., McCann, U., & Jesse, R. (2006). Psilocybin can occasion mystical-type experiences having substantial and sustained personal meaning and spiritual significance. Psychopharmacology, 187, 268–283.

[5] Poldrack, R. A. (2006). Can cognitive processes be inferred from neuroimaging data? Trends in Cognitive Sciences, 10(2), 59–63.

[6] Van Dam, N. T., van Vugt, M. K., Vago, D. R., Schmalzl, L., Saron, C. D., Olendzki, A., Meissner, T., Lazar, S. W., Kerr, C. E., Gorchov, J., Fox, K. C. R., Field, B. A., Britton, W. B., Brefczynski-Lewis, J. A., & Meyer, D. E. (2018). Mind the hype: A critical evaluation and prescriptive agenda for research on mindfulness and meditation. Perspectives on Psychological Science, 13(1), 36–61.

[7] Sutton, R. (2019). The Bitter Lesson.

[8] Ioannidis, J. P. A. (2005). Why most published research findings are false. PLoS Medicine, 2(8), e124.