Clara A. Sava-Segal

Note: Osher class members, this page has been updated after class to reflect the conversations we’ve had and some of the fresher materials have been moved to Lecture 3

Overview

In Session 1, we explored how expectations shape perception. Now we turn to a more fundamental question: Do different people experience the same sensory input in the same way? This session surveys subjective experiences beyond visual perception, focusing on dimensions that present particular measurement challenges. We'll examine interoception (sensing internal body states), temporal perception (how we experience time), synesthesia (cross-modal sensory experiences), aphantasia (absence of mental imagery), and taste perception (genetic variation in gustatory experience). Each domain reveals substantial individual differences and raises the question: How do we study something as private as subjective experience? Through behavioral tasks and neuroimaging studies, we'll see how researchers attempt to map the structure of consciousness across these diverse dimensions.

Interoception: Sensing the Body from Within

What Is Interoception?

While we're familiar with exteroceptive senses (vision, hearing, touch, taste, smell) that tell us about the external world, interoception refers to sensing the body's internal state. This includes awareness of heartbeat, breathing, hunger, thirst, temperature, pain from internal organs, and the need to urinate.

Why Interoception Matters: Interoceptive signals form the basis of emotional feelings. Different theories propose different relationships between bodily sensations, cognitive appraisal, and emotional experience.

Figure: Different theoretical models of how interoceptive signals relate to emotional experience

Measuring Interoceptive Accuracy: The Heartbeat Detection Task

Schandry (1981) developed the most widely used measure of interoceptive accuracy:

Heartbeat Counting Task:

Purpose: The task is a simple and widely used way to assess an individual's ability to perceive their internal bodily signals, specifically their heart rate
Participants silently count their heartbeats over specific intervals (25, 35, 45 seconds) without taking their pulse
Their reported count is compared to actual heartbeats measured by ECG
Accuracy score = 1 - (|recorded beats - counted beats| / recorded beats)

Key Finding: People vary dramatically—from near-perfect accuracy (0.90+) to almost no correlation (0.10) between reported and actual heartbeats.

Methodological Concerns: Körmendi et al. (2021) demonstrated that the Schandry task is sensitive to non-interoceptive (top-down) influences. Performance-related expectation was strongly related to task performance (β = .595, p < .001) even after controlling for physiological factors. People with high expectations may categorize vague sensations (such as attention-evoked sensations) as heartbeats, leading to inflated scores. This raises questions: Does the task measure true interoceptive sensitivity, or partly cognitive beliefs about one's heartbeat?

Neural Basis of Interoception

Critchley et al. (2004) used fMRI to identify brain regions involved in interoceptive awareness:

Anterior insula: Shows activation during heartbeat detection; gray matter volume correlates with interoceptive accuracy
Anterior cingulate cortex: Integrates interoceptive signals with emotional and cognitive processing
Somatosensory cortex: Represents body state information

Threat Anticipation and Interoception: Liu et al. (2021) demonstrated a whole-brain signature that specifically predicts anxious anticipation under uncertainty. Importantly, this neural pattern is not sensitive to predicting pain, general anticipation, or unspecific emotional and autonomic arousal—suggesting that the subjective experience of threat anticipation has a distinct neural basis tied to interoceptive processing.

Patient Study: Patients with insular cortex damage show impaired interoceptive accuracy while maintaining normal exteroceptive perception. One patient could see, hear, and feel external touch perfectly but couldn't detect their own heartbeat or accurately judge their emotional arousal.

Interoception and Mental Health

Discussion Point: If two people have different interoceptive accuracy, do they experience emotions differently? Is someone with high interoceptive accuracy having "more emotional" experiences, or just noticing bodily signals that others miss? And critically—are they accurately perceiving their body, or do their beliefs shape what they think they perceive?

Paulus & Stein (2010) proposed a model integrating interoception with mental health, though the relationships remain debated:

Anxiety disorders:People with anxiety might get confusing signals from their body (like a racing heart) and interpret them as dangerous. The problem isn't necessarily that they feel their body "too much," but that they expect threat and their beliefs amplify ambiguous body sensations.
Depression: Some research finds that people with depression are less accurate at sensing their body's signals, though studies show mixed results. It's possible that their beliefs about themselves interfere with how they interpret what their body is telling them.
Alexithymia: "No words for feelings"—people who struggle to identify their emotions might process body signals differently. But we don't know if they actually feel less, or if they just have trouble interpreting what they feel.
Eating disorders: People with eating disorders may process hunger and fullness signals differently and feel disconnected from their body. Again, we can't tell if their body sends different signals, or if their beliefs about food and their body change how they interpret normal signals.

A More Nuanced View: Depression and anxiety may not simply be problems with sensing the body accurately. Instead, they might involve a complicated interaction between: (1) what your body is actually signaling, (2) your beliefs about what those signals mean, and (3) your brain trying to make sense of confusing or inconsistent body signals. The key challenge is figuring out whether someone truly feels their heartbeat differently, or whether they just interpret the same sensations differently based on their beliefs and expectations.

Critchley, H. D., et al. (2004). Neural systems supporting interoceptive awareness. Nature Neuroscience, 7(2), 189-195.

Synesthesia: When Senses Blend

Defining Synesthesia

Synesthesia occurs when stimulation of one sensory pathway automatically triggers experiences in another sensory pathway. About 4% of people have some form of synesthesia, though many don't realize they're unusual until adulthood.

Types of Synesthesia

Common Forms:

Grapheme-color: Letters/numbers evoke colors ("A is red, B is blue")
Sound-color (chromesthesia): Sounds trigger visual colors
Lexical-gustatory: Words evoke tastes ("Derek tastes like earwax")
Number-form: Numbers occupy specific spatial locations
Mirror-touch: Seeing someone touched triggers tactile sensation on own body

Behavioral Validation

Ramachandran & Hubbard (2001) demonstrated that synesthesia is perceptual, not just associative:

The Embedded Figure Test (click to see the figures in the paper):

Display of many 5s with a few 2s scattered among them, forming a triangle
Non-synesthetes: Hard to spot the triangle quickly
Grapheme-color synesthetes: Immediately see the triangle because 2s and 5s are different colors

Implication: The synesthetic color experience happens automatically and early enough in processing to aid perceptual grouping—it's not just imagination or learned association.

Neural Basis: Cross-Activation Theory

Hubbard & Ramachandran (2005) proposed that synesthesia results from excess connectivity between normally separate brain regions:

Grapheme-color synesthesia: The brain area representing letter shapes (fusiform gyrus) is adjacent to color processing area (V4). Extra connections between them cause letters to activate color areas.
Developmental hypothesis: All infants may have dense cross-modal connections that are pruned during development. Synesthetes retain more connections.
Genetic component: Runs in families, suggesting hereditary factors in pruning process. Brang & Ramachandran (2011) discuss why the synesthesia gene might have survived evolution—synesthetes may have advantages in cognitive processing such as creativity, metaphorical thinking, and cross-domain problem solving.

What Synesthesia Teaches Us

Implications for Consciousness:

Sensory processing is more interconnected than traditional models suggest
Conscious experience can include genuine perceptual qualities without external stimulation
There's substantial variation in how different brains construct sensory experience
The line between perception and imagination is blurrier than we assume
Everyone may have somewhat different "perceptual worlds" due to neural wiring differences

Hubbard, E. M., & Ramachandran, V. S. (2005). Neurocognitive mechanisms of synesthesia. Neuron, 48(3), 509-520.

Aphantasia: The Absence of Mental Imagery

Discovering "Blind Imagination"

Zeman et al. (2015) described a remarkable case that revealed how differently people experience mental imagery:

Patient MX:

65-year-old man who lost ability to visualize after cardiac surgery
Could still recognize faces, navigate, draw from memory—all visual tasks intact
But couldn't voluntarily summon visual images: "I know what my wife looks like, but I can't see her in my mind"
This wasn't a memory problem—he knew what things looked like, he just couldn't visualize them
Tells us that successful performance in visual imagery and visual memory tasks can be dissociated from the phenomenal experience of visual imagery

Aphantasia in the General Population

Prevalence: About 2-5% of population reports little to no visual imagery
Discovery: Many don't realize they're different until adulthood—they assume "picture this" is metaphorical
Spectrum: Imagery vividness exists on a continuum from aphantasia (none) to hyperphantasia (extremely vivid)
Multi-sensory: Some aphantasics also lack auditory, gustatory, or other imagery modalities

Measuring Mental Imagery

Vividness of Visual Imagery Questionnaire (VVIQ):

Participants rate vividness of imagined scenes (1 = no image, 5 = perfectly clear and vivid)
Example: "Visualize a friend or relative. Consider the exact contour of their face, head, shoulders, and body."
Aphantasics consistently score 16 (minimum possible); hyperphantasics score 75-80 (maximum 80)

Zeman, A., et al. (2010). Loss of imagery phenomenology with intact visuo-spatial task performance: A case of 'blind imagination'.

Session 2: The Many Dimensions of Subjectivity