🧠 neuroscience

Offset analgesia and onset hyperalgesia: A comparison between women with non-suicidal self-injury and healthy women

This study utilizing a combined offset analgesia and onset hyperalgesia protocol with fMRI found that women with non-suicidal self-injury exhibit a weaker onset hyperalgesia response compared to healthy controls, providing partial support for the hypothesis of enhanced pain modulation in this population.

Original authors: Fust, J., Kastrati, G., Bjureberg, J., Jayaram-Lindström, N., Hellner, C., Kosek, E., Fransson, P., Jensen, K. B., Lalouni, M.

Published 2026-02-16

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Fust, J., Kastrati, G., Bjureberg, J., Jayaram-Lindström, N., Hellner, C., Kosek, E., Fransson, P., Jensen, K. B., Lalouni, M.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

The Big Picture: The "Pain Volume Knob"

Imagine your body's pain system is like a stereo system with a volume knob.

Turning the volume down is pain inhibition (making pain feel less intense).
Turning the volume up is pain facilitation (making pain feel more intense).

For a long time, scientists have noticed that people who engage in Non-Suicidal Self-Injury (NSSI)—like cutting or burning themselves without trying to die—seem to have a very low "volume" for pain. They often say, "I don't feel it," or they can handle much more pain than the average person.

The big question this study asked was: Why?
Is their "volume knob" broken so it's stuck on "low"? Or is their brain just incredibly good at turning the volume down (inhibiting pain) and failing to turn it up (facilitating pain) when it needs to?

The Experiment: The "Temperature Rollercoaster"

To test this, the researchers put 76 women (37 with a history of self-injury and 39 healthy women) inside an MRI machine. They used a special device to heat up a spot on the women's calves.

They didn't just heat it up and leave it there. They used a clever "rollercoaster" trick to test two different things:

The "Relief" Test (Offset Analgesia): Imagine the heat is steady and painful. Suddenly, the temperature drops just a tiny bit for a split second, then goes back up. Even though the heat is still high, your brain perceives a massive drop in pain. It's like a sudden gust of cool air on a hot day feeling way cooler than the actual temperature. This tests how well the brain can turn the volume down.
The "Surprise" Test (Onset Hyperalgesia): Imagine the heat is steady. Suddenly, the temperature spikes just a tiny bit, then goes back down. Even though the heat is still high, your brain perceives a massive spike in pain. It's like a sudden bump in the road feeling much worse than the road itself. This tests how well the brain can turn the volume up.

What They Found

The researchers had two main guesses (hypotheses):

Guess 1: The self-injury group would be super good at turning the volume down (strong "Relief" response).
Guess 2: The self-injury group would be bad at turning the volume up (weak "Surprise" response).

The Results:

The "Relief" Test (Turning Volume Down): No difference. Both groups felt the same amount of relief when the temperature dipped. The self-injury group wasn't "super-powered" at blocking pain in this specific way.
The "Surprise" Test (Turning Volume Up): Big difference. The healthy women felt a sharp spike in pain when the temperature jumped. The women with a history of self-injury? They barely noticed the spike. Their brains failed to turn the volume up.

The MRI Scan:
When looking at the brain scans, the healthy women's brains lit up like a Christmas tree in the "pain processing" area when the temperature spiked. The self-injury group's brains stayed much dimmer. It was as if their brain's alarm system was muted.

The Takeaway: A "Dull" Alarm System

Think of the pain system like a car's alarm system.

Healthy brains are like sensitive alarms: If you touch the car, it beeps loudly. If you scratch it, it screams. They react strongly to changes.
The brains in this study were like a broken alarm. When the temperature changed (the "scratch"), the alarm didn't go off. They didn't amplify the signal.

The study suggests that the reason these women feel less pain isn't necessarily because they are actively fighting the pain with super-strength (inhibition). Instead, it seems their brains are passive. They don't amplify the pain signals when something changes. They are "dull" to the spikes in sensation.

Why Does This Matter?

It explains the behavior: If your brain doesn't scream "OW!" when you get hurt, it's much easier to hurt yourself without feeling the usual fear or shock. The "barrier" to self-harm is lower.
It's not just about "toughness": It's not that these women are just "tough." Their internal volume knob for pain is physically wired differently.
Future Help: Understanding that their brains fail to amplify pain (rather than just blocking it) could help doctors create better treatments. Maybe we need therapies that help "wake up" their pain alarm system so they can feel pain normally again, which might reduce the urge to self-harm.

In short: The study found that women who self-injure don't necessarily have "super-powers" to block pain; rather, their brains are surprisingly quiet when pain tries to get louder, making the world feel less sharp and intense than it does for everyone else.

1. Problem Statement & Background

Individuals engaging in Non-Suicidal Self-Injury (NSSI) consistently exhibit hypoalgesia (reduced pain sensitivity), characterized by higher pain thresholds, higher pain tolerance, and lower reported pain intensity compared to the general population.

Theoretical Framework: It is hypothesized that this trait stems from hyper-effective pain modulation, specifically an enhanced ability to inhibit nociceptive signals (pain suppression) and a reduced ability to facilitate them.
The Gap: Empirical evidence regarding pain modulation in NSSI is inconsistent. While some studies suggest enhanced conditioned pain modulation (CPM), others show no difference or even weaker modulation. Furthermore, previous studies have not utilized Onset Hyperalgesia (OH) or combined it with neuroimaging to investigate these mechanisms.
Objective: To investigate whether women with NSSI demonstrate a propensity to inhibit nociceptive signals more strongly (via Offset Analgesia [OA]) and facilitate signals less (via Onset Hyperalgesia [OH]) compared to healthy controls, using a combined protocol and functional MRI (fMRI).

2. Methodology

Participants

Sample: 76 women aged 18–35.
- NSSI Group ( $n=37$ ): Engaged in self-injury on $\ge$ 5 days in the past year.
- Healthy Control (HC) Group ( $n=39$ ): No history of depression/anxiety treatment.
Exclusion Criteria: Chronic pain, autoimmune disorders, fMRI contraindications, and recent suicide attempts/plans.
fMRI Sub-sample: 34 NSSI and 39 HC participants provided usable fMRI data (some NSSI participants were excluded due to piercings, claustrophobia, or scanner issues).

Experimental Protocol

The study utilized a combined OA and OH thermal stimulation protocol inside a 3T GE MRI scanner.

Stimulation: Thermal stimuli applied to the left calf using a 30×30 mm thermode.
Calibration: Temperatures were individually calibrated to achieve a baseline "medium" pain rating of 4–6/10 on a 0–10 Numeric Rating Scale (NRS).
Trial Structure (3 Intervals):
1. T1 (0–6.5s): Ramp-up from 38°C to the calibrated medium temperature.
2. T2 (6.5–12s): The critical manipulation phase:
  - Control: Temperature held stable.
  - OA Condition: Temperature briefly increased by 2°C, then returned to medium.
  - OH Condition: Temperature briefly decreased by 2°C, then returned to medium.
3. T3 (12–33s): Return to medium temperature, then baseline.
Conditions: Each condition (OA, OH, Control) was repeated twice in randomized order.

Data Analysis

Behavioral: Pain ratings were analyzed using linear mixed-effects models (LME) in R. The primary metrics were the mean pain ratings during the last 12 seconds of stimulation (T3).
- OA Response: Calculated as the difference between OA and Control conditions.
- OH Response: Calculated as the difference between OH and Control conditions.
Neuroimaging (fMRI):
- Preprocessing: Performed via fMRIPrep (realignment, normalization to MNI space, 8mm smoothing).
- Statistical Modeling: First-level subtraction analyses ([OA - Control] and [OH - Control]) followed by second-level group comparisons (Two-sample t-tests).
- Regions of Interest (ROIs):
  - OA: Brainstem, dlPFC, rACC, Thalamus.
  - OH: S1, S2, Insula, ACC, Inferior Frontal Cortex.
- Thresholds: $p < .001$ uncorrected (cluster extent 20 voxels); significance set at $p < .05$ FWE-corrected.

3. Key Results

Behavioral Findings (Pain Ratings)

Offset Analgesia (OA):
- A significant OA effect was observed across all participants ($NRS -0.90, p=0.001$).
- No significant difference was found between the NSSI and HC groups regarding the magnitude of OA ( $p=0.10$ ).
Onset Hyperalgesia (OH):
- A significant OH effect was observed across all participants ($NRS 0.38, p < 0.001$).
- Significant Group Difference: The NSSI group displayed a weaker OH response compared to controls ($NRS -0.54, p < 0.001$). This indicates that women with NSSI facilitate nociceptive signals to a lesser degree than healthy women.
Correlation: A weak but significant positive correlation was found between OA and OH responses across all participants ( $r = 0.28, p = 0.01$ ), suggesting they may rely on partially distinct mechanisms.

Neuroimaging Findings (fMRI)

Offset Analgesia: No significant activation clusters were found for the OA response, either within groups or between groups, in the whole-brain or ROI analyses.
Onset Hyperalgesia:
- Main Effect: Significant activation was observed in the Primary Somatosensory Cortex (S1) contralateral to the stimulation site (left calf) during the OH condition compared to control ( $p_{FWE} = 0.032$ ).
- Group Differences: No significant clusters were found between NSSI and HC groups at the corrected threshold.
- Exploratory Analysis: At an uncorrected threshold ( $p < .05$ ), healthy controls showed greater activation increases in S1 and S2 during OH compared to the NSSI group, though this requires formal validation.

4. Key Contributions

First Clinical OH Study: This is the first study to investigate Onset Hyperalgesia in a clinical psychiatric population (NSSI).
Combined Protocol: It successfully demonstrates the utility of a combined OA/OH protocol to assess both pain inhibition and facilitation simultaneously within a single session.
Neural Correlates: It provides the first fMRI data on OH in humans, identifying S1 activation as a key correlate of the OH response.
Differentiation of Mechanisms: The weak correlation between OA and OH responses suggests that pain inhibition (offset) and pain facilitation (onset) may be governed by distinct neurobiological mechanisms, rather than a single "pain modulation" system.

5. Significance & Limitations

Significance

Partial Support for Hypothesis: The study supports the theory of hyper-effective pain modulation in NSSI, but specifically regarding pain facilitation (reduced OH) rather than pain inhibition (OA).
Clinical Implications: Understanding that NSSI involves a specific deficit in pain facilitation (rather than just enhanced inhibition) could refine pharmacological or behavioral interventions targeting the nociceptive spectrum.
Risk Factor Insight: The findings align with the theory that altered pain modulation (specifically reduced facilitation) contributes to the high pain tolerance observed in NSSI, potentially lowering the barrier to self-harm.

Limitations

Sample Size: The study was powered for pain threshold differences, not necessarily for subtle fMRI effect sizes, potentially leading to Type II errors (false negatives) in neuroimaging.
Control Group: The use of non-psychiatric controls makes it difficult to isolate whether the findings are specific to NSSI or related to comorbid psychiatric traits (e.g., depression/anxiety).
Protocol Variance: The OH protocol used differed slightly from the original Alter et al. (2020) study, which may have produced weaker effect sizes.
Null OA Results: The lack of significant OA brain activation clusters contradicts previous literature, suggesting a need for larger-scale studies to map OA neural circuitry accurately.

Conclusion

The study concludes that women with NSSI exhibit a reduced capacity to facilitate nociceptive signals (weaker Onset Hyperalgesia) compared to healthy controls, offering partial support for the hyper-effective pain modulation hypothesis. However, they do not show enhanced Offset Analgesia, suggesting that the hypoalgesic trait in NSSI may be driven more by deficits in pain facilitation than by superior pain inhibition.