Huberman LabEating for Better Sleep & Foods that Improve Metabolic Health | Dr. Marie-Pierre St-Onge
CHAPTERS
- 0:00 – 12:22
How sleep loss reshapes hunger signals and food reward (including sex differences)
Huberman and St-Onge frame sleep and nutrition as a two-way system: sleep affects what/how much we eat, and diet feeds back onto sleep quality. St-Onge describes controlled lab studies showing that sleep restriction increases overeating and alters appetite hormones differently in men versus women, alongside increased reward-center responses to food.
- •Population studies link short sleep with higher BMI and long-term weight gain, but don’t prove causality
- •In a 5-night lab sleep-restriction study, participants ate ~300 extra calories when sleep-restricted
- •Men: increased ghrelin (hunger signal); Women: reduced GLP-1 (satiety signal)
- •Sleep restriction increases brain reward responses to food cues, making palatable foods more compelling
- 12:22 – 17:06
Why sleep restriction leads to overeating: fatigue, thermic effect of food, and decision overload
They unpack why people tend to seek more palatable foods when tired, beyond simple willpower narratives. St-Onge highlights subconscious drives to eat for energy and alertness, plus reduced cognitive bandwidth for making “good choices.”
- •Sleepiness correlates with stronger drive for pleasurable foods
- •Eating can transiently increase alertness via the thermic effect of food
- •Fatigue reduces willingness to deliberate at the ‘buffet table’—defaulting to hyperpalatable options
- •Meta-analyses show ~250–400 kcal/day overeating with insufficient sleep
- 17:06 – 22:58
Metabolic health: severe lab sleep loss vs real-life mild chronic sleep loss
St-Onge contrasts a tightly controlled inpatient study (same foods, same timing) with a free-living long-duration sleep restriction study. Under strict control, short sleep alone didn’t worsen glucose/insulin/cortisol, but in real life, sustained modest sleep loss increased insulin resistance and blood pressure.
- •In-lab, isocaloric feeding can mask metabolic deterioration seen in the real world
- •5 nights of ~4 hours time-in-bed showed no cortisol, glucose, or insulin differences under control
- •A 6-week, ~1.5-hour/night reduction (to ~6 hours) increased insulin resistance and blood pressure
- •Post-menopausal women showed worse insulin sensitivity impacts than pre-menopausal women
- 22:58 – 28:35
Energy expenditure and NEAT under sleep loss: small increases don’t offset overeating
They explore whether being awake longer raises energy burn enough to counteract extra calories. St-Onge describes metabolic chamber findings: energy expenditure rises slightly because wakefulness costs more than sleep, but the increase is too small to offset typical sleep-loss overeating.
- •Metabolic chambers measure minute-by-minute O2/CO2 to estimate energy expenditure
- •Sleep restriction slightly increases daily energy expenditure (~5% or ~90 kcal in one study)
- •The typical overeating (~250–400 kcal) exceeds the extra burn, yielding positive energy balance
- •NEAT (spontaneous movement) can vary widely across individuals and meaningfully affect weight over time
- 28:35 – 35:04
Diet quality predicts future sleep: Mediterranean and DASH patterns vs insomnia risk
Switching direction, St-Onge summarizes longitudinal cohort analyses asking whether diet influences later sleep outcomes. Higher adherence to Mediterranean and DASH-style diets is associated with fewer insomnia symptoms and better odds of adequate sleep, even after adjusting for many covariates.
- •MESA cohort: Mediterranean-style diet linked to reduced insomnia symptoms and better sleep duration odds
- •Women’s Health Initiative: Mediterranean and DASH alignment predicted lower insomnia risk at 3-year follow-up
- •They model insomnia as dynamic (can resolve or develop), not a static baseline condition
- •Observational studies adjust for many confounds but can’t fully capture factors like social interaction
- 35:04 – 38:41
What you eat changes sleep architecture: fiber, saturated fat, refined carbs, and sleep depth
Using their inpatient dataset with polysomnography, St-Onge describes how self-selected diets changed sleep quality compared to controlled diets. Key findings link higher fiber with more slow-wave (deep) sleep, saturated fat with less deep sleep, and refined carbs with more nighttime arousals.
- •Self-selected diet differed: ~450 extra calories and ~33% more saturated fat than controlled feeding
- •Sleep duration didn’t change, but sleep latency increased (~70% longer) with self-selected intake
- •Deep slow-wave sleep was shorter (~20%+) when self-selecting vs controlled diet
- •Higher fiber predicted more deep sleep; saturated fat predicted less deep sleep; refined carbs predicted more arousals
- 38:41 – 40:34
Food timing for sleep: why earlier eating helps and why late meals can disrupt sleep
They discuss individual variability in tolerating late meals, but converge on earlier eating as generally better for sleep and cardiometabolic health. St-Onge shares her personal guideline of finishing food about three hours before bed, noting thermogenesis and body temperature as key mechanisms.
- •Late eating can impair sleep partly by increasing body heat (thermic effect of food)
- •St-Onge’s personal practice: last meal ~3 hours before bedtime
- •Controlled-feeding schedules vs ad-lib timing may explain differences in sleep outcomes
- •General principle: earlier eating tends to support better sleep and metabolic health
- 40:34 – 49:54
Chronotype, shift work, and naps: aligning behavior with your internal circadian clock
They address why the same sleep duration can feel different depending on bedtime, emphasizing circadian alignment. Shift work is framed as circadian mismatch, and naps are discussed as a tool that can help or harm depending on timing and whether underlying sleep disorders are present.
- •Perceived ‘hours before midnight’ effect likely reflects circadian misalignment when bedtime shifts
- •Shift workers struggle because sleep occurs when circadian biology promotes wakefulness (and vice versa)
- •Nap guidance: keep naps short (~30–60 minutes) and not too late to preserve sleep pressure
- •If daytime sleepiness persists despite adequate opportunity, investigate nighttime sleep quality issues
- 49:54 – 53:00
Women, sleep quality complaints, and clinician screening: multidimensional sleep health
St-Onge outlines known sex differences: women often sleep slightly longer yet report worse sleep quality and more insomnia symptoms. She argues clinicians should ask open-ended questions about sleep to uncover issues like snoring and sleep apnea, emphasizing that sleep health includes more than hours slept.
- •Women report more insomnia symptoms across adulthood despite slightly longer average sleep
- •Hormonal cycles, discomfort, and social roles may contribute to perceived poorer sleep
- •AHA scientific statement: clinicians should ask ‘How’s your sleep?’ rather than only sleep duration
- •Sleep health dimensions include regularity, satisfaction, daytime alertness, and nighttime disruptions
- 53:00 – 56:47
Snoring and sleep apnea: what to look for and how testing/treatment works
They clarify how snoring relates to apnea and why untreated apnea is a major health risk. St-Onge recommends formal testing (including at-home testing options) and explains why CPAP requires individualized pressure settings rather than being a one-size consumer purchase.
- •Common signs: loud snoring, gasping/arousals, unrefreshing sleep, daytime sleepiness
- •Apnea involves airway obstruction at the throat; nasal strips may reduce snoring but not fix apnea
- •Testing options include in-lab polysomnography and doctor-prescribed home sleep tests
- •Treatment often starts with weight loss if applicable; CPAP settings must be individualized for efficacy
- 56:47 – 1:00:30
Functional foods in the lab: kefir, fermented foods, and what ‘null results’ mean
Huberman asks about kefir and fermented foods; St-Onge recounts her early work testing kefir for cholesterol synthesis and finding no effect in that specific design. They broaden to the idea that functional foods may influence other endpoints (glycemic control, inflammation) even if one chosen outcome is null.
- •Kefir study: two cups/day vs milk in mildly hypercholesterolemic men showed no cholesterol-synthesis effect
- •A null result can reflect population, dose, comparator, or endpoint selection—not ‘no effects at all’
- •St-Onge supports fermented foods and the broader goal of supporting the gut microbiome
- •Discussion highlights how controlled trials can miss ‘side effects’ noticed anecdotally
- 1:00:30 – 1:11:28
Coffee mannooligosaccharides and ginger: small dietary tweaks that can change energy expenditure
They explore less-common nutrition interventions St-Onge studied: an industry-sponsored coffee-derived mannooligosaccharide product associated with improved body composition in men, and a ginger study showing an increased thermic effect of food. The broader theme is using modest, mechanistically grounded changes to ‘tip the scale’ over years.
- •Mannooligosaccharides were extracted from spent coffee grounds; effect observed in men, not women
- •Product wasn’t commercialized after sex-specific findings limited marketability
- •Ginger (powder in warm water) increased thermic effect of food over several hours
- •Functional-food research aims for small daily shifts that compound into long-term weight differences
- 1:11:28 – 1:17:20
Meal timing and fat oxidation: why earlier eating windows can support fat burning
St-Onge describes controlled metabolic chamber work where identical meals consumed later led to reduced fat oxidation. They compare an early (about 8am–6pm) versus late (about 12pm–10pm) 10-hour eating window, and connect findings to real-world data (e.g., earlier lunch improving weight loss in Spain).
- •Metabolic chamber study: same foods/amounts, different timing—later meals reduced fat oxidation
- •Early window example: ~1 hour after waking to ~10-hour window (e.g., 8am–6pm)
- •Late window example: delayed start by ~4 hours (e.g., 12pm–10pm)
- •Observational and intervention data suggest earlier meal timing supports weight management even in late-eating cultures
- 1:17:20 – 1:34:32
MCTs, body composition, and practical takeaways (plus the Portfolio Diet concept)
They revisit medium-chain triglycerides (MCTs): how they’re metabolized differently, produce a modest thermic boost, and can support weight loss when substituted for other fats. The discussion extends to additive ‘portfolio’ approaches—combining multiple evidence-based components for larger effects (e.g., LDL reduction comparable to statins).
- •MCTs (C8/C10) go more directly to the liver and are oxidized more readily than long-chain fats
- •Thermic effect increase is modest (~45–60 kcal per meal) but can matter when repeated daily
- •Weight-loss studies: MCT oil vs olive oil showed greater weight loss when used as a replacement, not an add-on
- •Portfolio Diet: soy/legumes, nuts, plant sterols, soluble fiber (later expanded) can reduce cholesterol substantially
- 1:34:32 – 1:41:19
Seed oils, processed foods, and reading labels: separating oil type from ultra-processing
They discuss an industry-funded snack study where corn-oil-fried chips improved lipid markers relative to other snack types, then widen to the seed oil controversy. St-Onge emphasizes smoke points and appropriate culinary use, and both highlight that ultra-processed food contexts can confound debates about specific fats.
- •Corn-oil-fried snack substitution improved lipid profile and lowered Lp(a) versus comparator snacks in a crossover design
- •Smoke point matters—some oils are unsuitable for high-heat cooking; olive oil isn’t ideal for deep frying
- •Public confusion often comes from seed oils being ubiquitous in ultra-processed foods
- •Practical tool: nutrition-label literacy and product comparison; personal needs (e.g., salt sensitivity) guide priorities
- 1:41:19 – 1:57:04
Industry-funded studies, null-result publication bias, supplements vs whole foods, and closing tools
They directly address concerns about industry sponsorship, emphasizing contractual right-to-publish and the broader scientific problem of journals rejecting null findings. They close with practical nutrition philosophy—prioritize whole foods, fiber and plant compounds—and the core ‘virtuous cycle’ message linking better sleep to better eating and vice versa.
- •Industry sponsorship doesn’t inherently dictate outcomes; misconduct can occur in any funding context
- •Major systemic issue: null results are hard to publish, creating a distorted literature
- •St-Onge prefers nutrients from whole foods; supplements may be appropriate for some individuals
- •Key framework: break the vicious cycle (poor sleep → poor food choices → poorer sleep) and build a virtuous cycle
