Social Baseline Theory and Perception: How our relationships ground us to our neurological starting point

An introduction to Social Baseline Theory and the Bayesian Brain
Social Baseline Theory is based on the principle that we consistently operate under an economy of action — that is, our actions and responses have a biological (and neurological) cost that causes a deviation from our baseline, which operates as our threshold of tolerance for certain stimuli. This can be seen in a basic homeostasis example where in response to heat, the body sweats to regulate its core temperature and we may decide to drink a cold glass of water. This is a simple example where our body engages in a mechanism to control our body’s temperature and restore it to baseline. We do not think about sweating, but we do think about what else we can do to return our body to its baseline state, as not only is the sensation of being overheated uncomfortable — but it is also our body telling us that overheating, which is a form of deviating too far from our baseline — can have dangerous consequences.

To apply this principle through a social baseline lens, we can draw upon evidence of mothers and the resource-demanding task of raising young children. For example, where child-rearing is played out on a traditionally gendered stage, women with male partners who are less active as parents often report increased stress and fatigue, and experience an increased chance of developing both mental and physical health issues. In contrast, parents who reported having an equal share of child-rearing duties reported happier relationships and fewer health issues. Likewise, the same can be said for single mothers who cohabitate with other single women to share the burden of raising children. In short, when tasked with overcoming a resource-heavy task, those who can share the load in equal parts with others experience less stress, increased health benefits and overall avoid the cost of having to navigate burden-heavy challenges alone. It is, in essence, the process of engaging in a cost-benefit analysis on both a subcortical and cortex level, and the Bayesian Brain theory can explain this process.

Example A: Mothers with less-active partners

Example B: Mothers with active partners — presence of equal parenting

The Bayesian Brain theory proposes that our mind operates as a cognitive decision-making machine which assesses and responds to new information to calculate the probability of an outcome — and the associated costs-benefits of that outcome — to inform our rationality and conscious decision-making. Further, the foundations of these decisions are based on our individual perceptions, learnings and memories, otherwise known as priors. A person’s priors are unique to the lived experiences of that individual and provide the foundational data on which the Bayesian Brain perceives information of current environmental and situational changes to assess the available resources that person has to manage and respond to those changes. In essence, the purpose of the Bayesian Brain is to determine the energy output required by an individual in response to a predicted outcome and to reach a decision that will cost the least resources. This predictive process encompasses a cognitive algorithm which considers the following:

1. The constraints, risks and opportunities of the current situation;

2. the predicted possible future situation(s) of each respective outcome;

3. situational goals of the individual;

4. current energy levels; and

5. expected future energy levels.

The Bayesian Brain theorem has been applied to machine learning and has allowed AI to mimic decision-making similar to humans — thus supporting the theory that the human brain is Bayesian. Likewise, this algorithm of decision-making shares similarities with Scherer’s Appraisal Theory which is also informed by a person’s lived experiences. Overall, Social Baseline Theory relies on Bayesian Brain processes and is a hypothesis that is rooted deeply in how individuals perceive the world. These perceptions are drawn from external stimuli and our stored memories, knowledge and inferences of the world based on past experiences. It is, in essence, a theory that proposes that the nature of the social relationships we come into contact with informs the way we perceive the world — so where positive social relationships exist, we are to perceive the challenges as less burdensome. In contrast, the opposite can be said for the presence of negative relationships. Social Baseline Theory, with its foundation in the Bayesian Brain theorem, is essentially a hypothesis based on perception.

An argument against top-down effects on perception and a response
Schnall et al.’s 2008 study tested the theory that our visual perception of a physical challenge is influenced by the resources available to us.[1] Two studies were conducted, which required participants to measure the steepness of a hill. In the first study, participants were supported by a friend, while the second study paired participants with someone they disliked. In the first, it was found that participants perceived the hill to be less challenging to overcome and therefore less steep than the hill’s actual angle. In contrast, participants in the second study reported that they felt the hill was more challenging and would require more physical exertion. These findings conclude that the presence of social support leads individuals to mentally perceive challenges as less resource-demanding, where the company of one or more persons allows the group to share the load of the perceived challenge. This occurrence of perceiving a potential challenge to be less burdensome based on an individual’s internal assessments such as expectations, emotional and mental resources and physical capabilities is known as top-down processing.

However, Firestone and Scholl have argued against the top-down effects of perception. Their argument engages six elements that seek to explain the theoretical deficit of top-down cognitive perception theories. Accordingly, these deficits are:

1. Most top-down hypotheses are tested only through confirmatory predictions and fail to account for disconfirmatory predictions. Firestone and Scholl draw this conclusion based on the El Greco fallacy.

2. Most top-down theories rely on empirical evidence that is based on inferences of perceptual reports, which are informed by our visual data — and not judgement-based explanations.

3. The nature of experiments, particularly psychology experiments that feature an element of social interaction, can lead to a conscious or unconscious assumption about the experiment, which will result in subjects adjusting their responses — in essence, they do not account for the presence of demand bias.

4. Low-level differences in stimuli are not accounted for in experiments that test top-down effects on perception. An example of low-level differences may be the differences in the colour and size of a spider.

5. Studies that focus on visual perception do not adequately determine if the perceived top-down state of being (e.g. an emotion such as anger, fear etc.) influences what we directly see or if peripheral vision is encompassed in identifying attentional effects.

6. Top-down effects on perception are informed by what individuals see. However, most studies focus on the effects of recognising certain stimuli. Accordingly, the ‘back end’ memory phenomena cannot explain the top-down effects on visual perception.

Although Firestone and Scholl may successfully apply these arguments to situations that rely solely on visual perception, their claims fall short in their application to perception as a whole. First, the El Greco fallacy presents the problem that this abnormality will be present in similar interactions where visual perception abnormalities exist. However, while the El Greco fallacy may account for visual perception, it fails to account for the other elements that account for perception and cannot be applied to perception as a whole. Second, Firestone and Scholl suggest that subjective perceptual reports should be included in experimental hypotheses which test perception. They also argue against the term ‘perceptual judgement’ in various literature but fail to provide an alternative suggestion or clear argument as to why this term is inappropriate. Third, regarding the presence of demand biases in psychological experiments, there seems to be a failure to account for the fact that most experiments do account for the existence of demand bias in experiments, and this method of accountability is widely practised within the profession. To apply their argument to all experiments seems to be a misleadingly broad categorisation informed by a misunderstanding of how demand characteristics are accounted for and controlled in psychological experiments. Fourth, Firestone and Scholl appear to make the same error made with their third claim. Yet, while they argue that low-level differences may not be adequately addressed in top-down studies, they fail to provide evidence or a supportive rationale that low-level differences would result in radical variations of top-down perceptions. Fifth, there is a large body of research that shows perception is not separate from visual stimuli and that visual perception can be influenced by a person’s emotional state of being. Likewise, studies that do measure visual attention, also take measures to account for peripheral influences, if they are deemed relevant. Last, the predominant argument that Firestone and Scholl appear to argue is that there is a lack of clear distinction between perception which is an outcome of cognitive thought and perception which is an outcome of visual sensory input. Their reasoning seems to be that without a conscious shift in perception, the result of whichever perceptional processes have been engaged, does not constitute what most would categorise as perception. However, while the differences between thought and visual perception may be products of varying cortical or sub-cortical processes, their theory nonetheless fails to account for the fact that perception is not a stand-alone function that operates in isolation and neither it is influenced solely by visual input. Instead, perception is the process by which one draws information from both external sensory data and internal knowledge, experiences and inferences about the world.

If our perception is informed by the data, memories and knowledge we’ve accumulated throughout our lifetimes, it would be safe to conclude that this data constitutes as our priors. The Bayesian Brain then calculates these priors to form a perception of the world based not only on visual perception, but on the circumstances as a whole and the cost-benefit of the actions involved in responding to the situation in which a person finds themselves. If the Bayesian Brain principle is plausible, to which there is ample evidence available to support this conclusion — then it would not be difficult to conclude that the presence of social resources would also be calculated as either beneficial or detrimental to a given circumstance. When applying the Bayesian Brain theorem, which relies on the principle that perception encompasses more than just visual stimuli, it is reasonable to provide that the Social Baseline Theorem has merit. However, it is not just evidence of this cost-benefits analysis on homeostasis resources. The presence of neurological benefits also provides an additional element. As mentioned above, when social support is present, the human brain declines in activity. However, considering that Social Baseline Theory is a reasonably new theorem, there undoubtedly is a need for further research that introduces social support and conducts studies with the assumption that social baseline is the starting line for processes.

In discussing the feasibility of Social Baseline Theory, it is necessary to address the principle of the Bayesian Brain and the analysis of the costs-benefits involved in making assessments of perception. Firestone and Scholl have argued that perception cannot be based on top-down processes on which the Bayesian Brain relies. However, their argument fails to hold merit when applied to theories that fall outside the scope of visual perception and account for the multi-faceted knowledge, inferences and lived experiences that inform people’s perceptions and by extension, their decisions. Nevertheless, if the Bayesian Brain theorem is sound, and there is evidence available that provides that social resources act as an anaesthetises to stress and counter-balances resource-heavy challenges, then it would be reasonable to provide that Social Baseline Theory is plausible.


Agnafors, Sara et al, ‘Mental Health in young mothers, single mothers and their children’ (2019) 19 (112) BMC Psychiatry

Baker Lewis J. and Daniel T. Levin, ‘Face-Race Lightness Illusion Is Not Driven by Low-level Stimulus Properties: An Empirical Reply to Firestone and Scholl’ (2016) 23 Psychonomic Bulletin & Review 1989–1995

Coan, James A. and David A. Sbarra, ‘Social Baseline Theory: the social regulation of risk and effort’ (2015) 1(87) Current Opinion in Psychology 87–91

Coan, James A., Hillary S. Schaefer and Richard J. Davidson, ‘Lending a hand: Social regulation of the neural response to threat’ (2006) 17(12) Psychological Sciences 1032–1039

Coan, James A. ‘Towards a Neuroscience of Attachment’ in Jude Cassidy and Philip R. Shaver (eds), From The Handbook of Attachment: Theory, Research and Clinical Implications (The Guilford Press, 2008) 241–265

Coan, James A. and Lane Beckes, ‘ Social Baseline Theory: The Role of Social Proximity in Emotion and Economy of Action’ (2011) 5(12) Social and Personality Psychology Compass 976–988

Eldermire, Summer ‘The single mums who live together on ‘mommunes’’, BBC (11 September 2019)

Firestone, Chaz and Brian J. Scholl, ‘Cognition does not affect perception: Evaluating the evidence for “top down” effects’ (2016) 39 Behavioural and Brain Sciences 1–77

Firestone, Chaz and Brian J. Scholl, ‘”Top-Down” Effects Where None Should Be Found: The El Greco Fallacy in Perception Research” (2014) 25(1) Psychological Science 38–46

Gantman, Ana P. and Jay J. Van Bavel, ‘Commentary on Firestone & Scholl: Behaviour is multiply determined and perception has multiple components: The case of moral perception’ 2016) 39 Behavioral and Brain Sciences

Ghahramani, Zoubin, ‘Probabilistic machine learning and artificial intelligence’ (2015) 521 Nature 452–459

Gross, Elizabeth B. and Sara E. Medina-Devilliers, ‘Cognitive Processes Unfold in a Social Context: A Review and Extension of Social Baseline Theory’ (2020) 11(378) Frontiers in Psychology 1–12

Manjaly, Zina-Mary and Sandra Iglesias, ‘Computational Theory of Mindfulness Based Cognitive Therapy from the “Bayesian Brain” Perspective’ (2020) 15(11) Front Psychiatry

McDonnell, Cadhla, Nancy Luke and Susan E. Short, ‘Happy Moms, Happier Dads: Gendered Caregiving and Parents’ Affect’ (2019) 40(17) Journal of Family Issues 2553–2581

Orne, Martin T ‘Demand Characteristics and the Concept of Quasi-Controls’ in Robert Rosenthal and Ralph L. Rosnow (eds) ‘Artifacts in Behavioural Research’ ( Oxford University press, 2009) 110–150.

Scherer, Klaus, ‘The dynamic architecture of emotion: evidence for the component process model’ (2009) 23(7) Cognition and Emotion 1307–1351

Schnall, Simone, ‘Social Support and the Perception of Geographical Slant’ (2008) 44(5) Journal of Experimental Social Psychology 1246–1255

Schnall, Simone ,‘No Magic Bullet in Sight: A Reply to Firestone and Scholl (2017) and Durgin (2017)’ (2017) 12(2) Perspectives on Psychological Science 347–349

Summerfield Chirstopher and Tobias Egner, ‘Expectation (and attention) in visual cognition. Trends in Cognitive Sciences’ (2009) 13 (9) Trends in Cognitive Sciences 403–409



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Laura E Fox

LLB (Hons) and BA (Gender Studies and Philosophy) student. A collection of academic essays.