Everyday life is filled with decisions involving combinations of things: a lunch meal, a phone plus a data plan, a vacation package, or a streaming service bundle. While these choices seem straightforward, the brain's method for evaluating multi-component options has remained surprisingly mysterious. A new study published in the Journal of Neuroscience offers rare insight into what happens inside the human valuation system when single items become bundles - and why the total often feels like less than the sum of its parts.
The researchers conducted a three-day deep-fMRI experiment with 14 participants who assigned monetary values to individual consumer items and to bundles containing two items from the same set. Behaviorally, the results showed a consistent pattern: people valued bundles sub-additively, meaning the combined value of two items was noticeably lower than simply adding their separate values. This effect has been documented in behavioral economics for years, but the neural basis behind it has been unclear.
Inside the scanner, a clearer picture emerged. The researchers found that the ventromedial prefrontal cortex (vmPFC) - a region known to encode subjective value across rewards, goods and outcomes - processed both single-item and bundle valuations. The same was true for a broader network across the prefrontal cortex. These regions acted as a "general value code" independent of whether the choice involved one item or many.
But the value representation did not stay constant across contexts. When participants evaluated bundles, vmPFC activity appeared attenuated, as though the brain shifted into a different scaling mode. Instead of treating each added item as an independent contribution, the neural signal compressed the combined value. The authors interpret this as evidence that the brain uses relative, normalized value rather than an absolute coding system. In other words, the brain adjusts how strongly it responds based on the context - including the complexity or number of components involved.
This type of normalization is common in sensory systems: neurons scale their responses based on the distribution of incoming stimuli, ensuring no signal overwhelms the system. Seeing something bright in a dark room activates visual neurons strongly; seeing the same brightness in daylight produces a far smaller response. The current study suggests that the valuation system uses a similar rule. When multiple items occupy the same mental space, the brain adjusts its representation so that the whole does not exceed a manageable neural range.
Such scaling may help make decisions more efficient. Evaluating multi-component objects could otherwise overload the brain with redundant information. By compressing value signals, the system maintains a stable frame of reference even as choices become more complex. But this comes with a tradeoff: bundles may consistently feel less exciting or less valuable than their separate parts. This helps explain a wide range of consumer patterns, including why people sometimes undervalue meal combos, struggle to evaluate subscription plans, or feel uncertain about packages that mix several benefits.
The findings also support earlier evidence suggesting that the brain relies on a "common currency" of value, but one that is flexible rather than fixed. The valuation network can integrate disparate items under a unified scale, yet this scale dynamically adjusts depending on what else is present. Similar patterns have been observed in reward learning, choice overload, and even self-evaluation, where the presence of multiple factors dampens the overall signal.
Although the study included a small sample, the experimental design - pairing behavioral valuations with multiday fMRI scans - provides unusually detailed insight into how value is computed. The authors argue that this normalization process may be a fundamental principle of human decision-making, especially when options become multi-dimensional.
The implications extend beyond shopping behavior. Human decisions frequently involve bundles in a broader sense: choosing between jobs that combine salary and culture, relationships that blend traits, or lifestyle decisions that merge tradeoffs and benefits. If the brain naturally compresses value when items combine, this could affect how people weigh complex life choices, potentially favoring simpler options that are easier to compute.
The study also touches on a deeper point: the brain does not operate according to strict arithmetic. Value is not a fixed quantity attached to objects, but a dynamic signal shaped by context. Even when people consciously believe they are adding things up, the valuation system may be quietly rescaling everything in the background.
Seen through the lens of Seven Reflections' Dimensional Systems Architecture (DSA), this finding illustrates how cognitive fields regulate internal signals. When multiple components enter a single valuation field, the field does not expand linearly. Instead, it redistributes intensity to maintain coherence, leading to normalization. The brain is not failing to add correctly; it is stabilizing the field to preserve functional structure. Sub-additive valuation is an emergent property of field-level computation - the cognitive system protecting itself from overload by compressing incoming signals.
From a DSA perspective, the brain's valuation network behaves like a dynamic field that adapts to the density of cognitive elements. Singular items occupy a clear, high-amplitude point in the field. When several items occupy the same space, the field integrates them not by summation but by normalization, allowing the system to maintain balance. This aligns with the broader DSA principle that meaning emerges from structure, not from linear accumulation.
As research continues to explore how humans make complex decisions, this study provides a valuable window into the architecture of value itself. Choices are not simply computed; they are shaped by the interplay of internal structures that adjust to context.
