The brain is not designed to pay attention all the time. On the contrary, it is designed to notice what changes — and to ignore everything else as efficiently as possible.

The default state of the brain is not receptivity. It is filtering — a continuous, largely unconscious process of deciding what to ignore. The mundane, the predictable, the information that confirms what is already known: these pass through without registering. What breaks through is the novel, the surprising, the emotionally charged, the thing that doesn't fit the pattern the brain was expecting. This is not a limitation. This is the system working as designed — a cognitive architecture shaped by the need to detect what matters in an environment full of noise to save the energy and use it only when it’s important.

The methodological consequence is direct — and it operates upstream of analysis, at the point where research is designed. If the brain ignores the predictable, a stimulus that feels familiar will produce a response to familiarity rather than to the thing being tested. A question that signals its expected answer will receive that answer. A concept that looks like every other concept activates recognition rather than genuine engagement — and recognition is not the same as response. What you measure in that situation is the brain's filtering system, not the thing you were trying to reach.

This shapes what to test and how. The choice of stimulus, the framing of a question, the sequence in which ideas are introduced — all of these are decisions about what kind of attention you are inviting. Novel, unexpected, emotionally resonant material activates a different cognitive mode than the familiar and predictable. If the research question requires genuine reflection rather than habitual response, the design needs to create the conditions for that reflection to occur — which means building in the kind of surprise or dissonance that breaks the brain's default filtering before asking it to engage.

The indicators you choose to measure need to correspond to the cognitive process actually operating in the situation. Asking people to evaluate something they have barely processed, or to articulate a response to something that activated only surface recognition, produces data that is clean and largely meaningless. Knowing which process is running — and designing for the one you actually want — is what separates research that reaches genuine response from research that measures the performance of response.

Medina's observation that vision dominates sensory processing points toward a methodological challenge that is considerably more complex than it first appears. The problem is not that research relies too little on the visual — images, photographs, and concept boards are everywhere in contemporary practice. The problem is that their presence is rarely examined with the rigour applied to other research instruments.

A visual is not a delivery mechanism for a concept. It is a stimulus in its own right — and a remarkably dense one. Where a word carries connotations that vary by context and culture, an image multiplies those connotations across every element it contains: the faces, the light, the objects, the spatial relationships, the things that are present and the things that are (conspicuously) absent. Each viewer brings to it a different set of associations structured by culture, class, generation, and personal history — associations that are not random but are not shared, and that activate responses the research was not designed to measure. What Gombrich and Kandel established about perception applies with particular force here: there is no innocent image. Every visual is already interpreted before it is consciously processed, and the interpretation varies in ways that are too complex to easily untangle.

The practical consequence is that a research stimulus containing a visual is producing responses to considerably more than the intended concept. The associations travel with the image whether or not anyone invited them. This does not mean avoiding the visual — we are, as Medina reminds us, creatures for whom seeing is the dominant mode of making sense of the world. It means treating every image as a deliberate choice rather than a default, examining what it carries before deploying it, considering carefully where it deepens understanding and where it overstimulates without illuminating. The visual used sparingly and consciously is one of the most powerful tools in research design. The visual used as decoration, as reassurance, or as a substitute for an idea that hasn't quite been found yet, adds noise to a process that already has enough.