Could the placebo effect be bullshit? (with Literal Banana)

Jan 15, 2025 Episode Page ↗
Overview

Spencer Greenberg and Literal Banana discuss whether the placebo effect is real, examining evidence against it, the replication crisis, misattribution, and animal studies. They also explore arguments for its existence, like the impact of mindset and psychosomatic effects.

At a Glance
16 Insights
1h 27m Duration
14 Topics
9 Concepts

Deep Dive Analysis

Introduction to Placebo Effect Skepticism

Defining Placebo Effect vs. Dummy in Trials

Initial Theories on Placebo Group Improvement

Evidence Against a Strong Placebo Effect: Meta-Analyses

Limitations of Placebo vs. No-Treatment Study Design

Understanding Effect Sizes and Cohen's D

Critique of Meta-Analysis Quality and Small Trials

Placebo-Focused Studies and Exaggerated Effects

Arguments for a Placebo Effect: Belief, Misattribution, Mindset

The Role of Mindset in Pain Perception and Suffering

The Nocebo Effect and Psychosomatic Experiences

Animal Studies on Placebo Effect and Conditioning

Brain Imaging Evidence and Response Bias

Listener Reactions and Final Takeaways

Placebo Effect

The idea that an inert substance or practice, like a sugar pill or sham surgery, can produce a healing effect, often attributed to suggestion, belief, or the ritual of treatment. Literal Banana argues this effect is largely 'bullshit'.

Dummy (in RCTs)

An inert substance or procedure used in randomized controlled trials to mimic the active treatment, primarily to ensure blinding of participants and researchers. It serves as a control for non-specific effects, distinct from the concept of the placebo effect as a healing power.

Open-Label Placebo

A placebo treatment administered to a patient who is fully aware that they are receiving an inert substance. Some studies claim these can still be effective, challenging the traditional view that belief in the active substance is necessary.

Hawthorne Effect

The phenomenon where individuals modify their behavior or performance in response to their awareness of being observed or studied. This can lead to improvements in outcomes that are not due to the intervention itself.

John Henry Effect

The tendency for individuals in a control group to work harder or perform better than usual, specifically because they know they are in the control group and want to demonstrate their capability or compete with the experimental group.

Cohen's D (Standardized Mean Difference)

A unitless measure of effect size used to quantify the difference between two group means relative to the standard deviation of the data. It allows for combining results from studies that use different measurement scales in meta-analyses.

Stress-Induced Analgesia

A phenomenon where people report experiencing less pain after being subjected to a stressful situation, such as public speaking or difficult math problems. This suggests that stress can alter pain perception.

Psychosomatic Effects

Physical symptoms or changes in the body that are caused or influenced by mental states, emotions, or psychological factors. These are acknowledged as real and distinct from the placebo effect, as emotions can genuinely alter physiological responses.

Neurological Pain Signature (NPS)

A specific pattern of brain activity, detectable via fMRI, that is associated with the basic, nociceptive perception of physical pain, regardless of the type of painful stimulus. Placebos generally do not activate this signature.

?
What is the placebo effect?

It is defined as an inert substance or practice that produces a healing effect, often attributed to suggestion or the power of the mind, but its exact mechanisms are debated, with Literal Banana expressing strong skepticism about its actual healing power.

?
How does a placebo differ from a dummy in clinical trials?

A 'dummy' is an inert substance used in randomized controlled trials to blind participants and researchers to the active treatment, serving as a control. A 'placebo' (in the context of the placebo effect) is thought to have actual healing power through suggestion or other psychological means.

?
Why is it important to have both placebo and no-treatment groups in a study?

Comparing a placebo group to a no-treatment group helps isolate the specific effect of the placebo itself, distinguishing it from natural improvement over time, regression to the mean, or other non-specific factors that cause people to feel better.

?
What is the John Henry effect?

It is the tendency for individuals in a control group to exert extra effort or perform better than usual, possibly due to a desire to compete with or outperform the experimental group, despite not receiving the experimental treatment.

?
How large is the placebo effect when compared to no treatment?

Meta-analyses comparing placebo to no-treatment groups generally find a very small, statistically significant effect, primarily on self-reported continuous outcomes (e.g., pain scales), but not on objective measures or binary outcomes.

?
Can belief alone cause a placebo effect?

While belief and expectation are often cited, some research suggests the placebo effect might be independent of conscious belief, and the observed effects could be due to factors like politeness, role-playing, or response bias.

?
What do animal studies reveal about the placebo effect?

Animal studies attempt to induce placebo effects through conditioning (e.g., associating a stimulus with a pain reliever). However, these studies are difficult to replicate, and higher-quality studies often fail to find a significant or consistent placebo effect.

?
Does brain imaging support the existence of a placebo effect?

Brain imaging studies have shown no effect of placebos on the 'neurological pain signature' (basic pain perception). Some EEG measures related to disgust or distress show a placebo effect, but these appear to be under voluntary control, potentially reflecting response bias rather than genuine physiological change.

1. Mindfulness for Enhanced Pleasure

Practice mindfulness and gratitude during enjoyable everyday experiences, such as drinking tea, to significantly enhance the pleasure derived from them. Being in a certain mental state can double or triple the pleasure from simple activities.

2. Prioritize Truth Over Belief

Prioritize believing the truth, even if it means potentially foregoing a perceived placebo benefit. Believing the truth is considered most important, and evidence should be provided to help people believe true things.

3. Acknowledge Psychosomatic Effects

Understand that strong emotions, particularly anxiety, can genuinely manifest as physical symptoms in the body. Anxiety hormones and chemicals can course through the body and create real physical sensations and illness.

4. Overcome Post-Injury Adaptations

If you’ve recovered from an injury but still experience pain or stiffness due to ‘babying’ the affected area, consider challenging those learned adaptations and pushing through the discomfort. Unnecessary adaptations can perpetuate feelings of pain and stiffness even after the initial injury has healed.

5. Use Metaphors as Tools

Employ seemingly ‘woo’ or metaphorical concepts (e.g., visualizing energy flow) as practical tools to improve physical performance or body awareness, without believing them as literal metaphysical truths. These concepts can be useful frameworks for achieving desired physical states or motions, even if not literally true.

6. Conduct Self-Experiments

Regularly conduct self-experiments to test hypotheses about what works for you, even if many attempts don’t yield positive results. This is a way to learn about your own body and behaviors, despite a high rate of non-working interventions.

7. Seek Information to Reduce Anxiety

When experiencing anxiety, especially due to uncertainty or lack of information, actively seek trustworthy and convincing information to calm down. Anxiety can stem from a lack of information, and receiving reassuring, credible information can alleviate it.

8. Avoid Post Hoc Fallacy

Be aware of the ‘post hoc, ergo propter hoc’ fallacy, where one attributes an improvement to a treatment simply because it followed the treatment. Many conditions improve naturally or fluctuate, and attributing recovery solely to a treatment without proper evidence can be a misattribution.

9. Skepticism Towards Automaticity

Be skeptical of the idea that humans are automatons controlled by subtle environmental changes (automaticity) in psychological research. Many such effects have failed to replicate, suggesting they might be ‘fake’.

10. Critique Mixed-Quality Meta-Analyses

When interpreting meta-analyses, be critical of studies that combine low-quality research with high-quality research, as this can obscure the truth. Including ‘garbage’ trials can lead to misleading overall effect sizes and make it harder to discern reliable findings.

11. Avoid Small Sample Trials

When evaluating research, be skeptical of trials with small sample sizes (e.g., less than 100 people per group) as they are unlikely to provide much reliable information. Small sample sizes can lead to ‘garbage’ results and make it difficult to learn much.

12. Scrutinize Small Standard Deviations

When reviewing research, be critical of studies reporting unusually large effect sizes, especially if accompanied by implausibly small standard deviations. A small standard deviation can artificially inflate a standardized effect size, making it appear larger than it truly is.

13. Recognize Hawthorne Effect

Be aware that being observed or studied can alter behavior, leading people to act differently than they normally would. This effect can influence outcomes in studies or situations where observation is present.

14. Understand John Henry Effect

Understand that control groups might exert extra effort to prove themselves, potentially biasing results. This effect, known as the John Henry effect, can lead control groups to perform better than expected.

15. Continue Placebo-Controlled Trials

Continue using placebo-controlled trials in research, as they are one of the best ways to get accurate information about the effectiveness of treatments. This method helps differentiate true treatment effects from natural improvement and other factors.

16. Monitor External Eating Cues

Be aware that external cues, like the perceived fullness of a bowl, can influence how much you eat, overriding internal satiety signals. People tend to eat more if their bowl never appears to empty, potentially leading to overeating.

Is the placebo effect bullshit? I really think it is.

Literal Banana

I think placebo controlled trials are great and should continue and is one of the best ways to get information about treatments. But I don't have much hope for a powerful placebo effect.

Literal Banana

To me, the core of it is role-playing and politeness, that the placebo group may report that they're doing a little bit better because they've been given this communication that, you know, well, you're given the substance and it's supposed to make you better.

Literal Banana

If you average those together, you find a very, very small difference in favor of the placebo, but it's very small. So this suggests that a lot of the driving of that, you know, difference in the, from the, of the placebo actually working is actually coming from the tiny trials, really small trials.

Spencer Greenberg

I think it's probably good to have hope, but I don't know if I'd call that a placebo effect.

Literal Banana

I think our sensations in our body are a lot weirder than a lot of people realize, because very few people spend the time to, like, pay attention to them.

Spencer Greenberg

I would say zero healing effect and social desirability or demand characteristics responding.

Literal Banana

Animal Conditioning for Placebo Effect (Morphine)

Literal Banana
  1. Expose the rat to a painful stimulus when it has been administered morphine to create an association.
  2. Repeat this conditioning process.
  3. Later, expose the rat to the same stimulus but administer a saline injection (no morphine), expecting the rat to exhibit fewer pain behaviors due to the learned association.

Animal Conditioning for Placebo Effect (Housing Unit)

Literal Banana
  1. Condition rats to associate a particular housing unit (with specific visual, smell, or sound cues) with low pain by applying low-level painful stimuli in that environment.
  2. On an experiment day, increase the painful stimulus in that same housing unit, expecting the rats to experience less pain or exhibit fewer pain behaviors due to their expectation of low pain in that environment.

Human Conditioning for Placebo Effect (Electric Shocks)

Literal Banana
  1. Train participants by giving them lower-level, non-painful electric shocks when a 'green light' is on, and bad shocks when the 'green light' is off.
  2. Ask participants to rate their pain throughout, expecting them to rate the 'green light' shocks lower.
  3. In a test phase, administer the same level of shock regardless of whether the 'green light' is on or off, observing if participants initially report lower pain with the 'green light' due to conditioning before realizing the change.
3 to 6 points
Placebo effect on pain (100-point scale) Better than no treatment, found in meta-analyses of pain treatment trials comparing placebo to no treatment.
0.26
Placebo effect on patient-reported continuous outcomes (Cohen's D) Effect size relative to no treatment, based on the 2010 Hróbjartsson and Gøtzsche meta-analysis.
0.13
Placebo effect on observer-reported continuous outcomes (Cohen's D) Effect size relative to no treatment, based on the 2010 Hróbjartsson and Gøtzsche meta-analysis.
0.93
Placebo effect on binary outcomes (Relative Risk) Pooled effect, meaning the placebo group was 7% less likely to report a bad outcome compared to no treatment, with a 95% confidence interval of 0.88 to 0.99.
6%
Percentage of people believing in no less pain from placebo Result of a Twitter poll conducted by Spencer Greenberg asking about belief in a placebo's effect on chronic pain.
41%
Percentage of people believing in slightly less pain from placebo Most common answer in a Twitter poll conducted by Spencer Greenberg about belief in a placebo's effect on chronic pain.
1.73
Highest Cohen's D reported in some placebo-focused meta-analyses Reported in studies specifically focused on the placebo effect, with some individual studies claiming effects over 7.