The popular idea that we “get used to” anything is only half true. The panel evidence shows adaptation is partial, uneven, and specific to the injury.
When someone is seriously hurt, their happiness falls, and any damages proceeding must eventually ask whether it comes back. For years a popular idea in psychology, the "hedonic treadmill," held that people adapt to almost anything and drift back to a fixed personal baseline no matter what befalls them. If that were true, pain-and-suffering losses would be mostly temporary. The modern evidence, built on large multi-year panel studies that follow the same people before and after an event, tells a more precise and more useful story: adaptation is real but partial, uneven, and specific to the type of loss. People recover well from some events, poorly from others, and from a few barely at all. This review sets out that evidence, and what it implies for valuing damages over time.
The idea traces to Brickman and Campbell's 1971 essay on "hedonic relativism," which argued that pleasure and pain are judged against a moving baseline of prior experience, so gains and losses fade. The empirical touchstone is Brickman et al. 1978, comparing recent lottery winners, people paralyzed in accidents, and controls. It is widely misremembered as showing that the accident victims "adapted back." They did not: the study reported that paralyzed victims were significantly less happy than controls and took less pleasure in everyday activities, even as they were happier than one might naively fear. Even this small founding study showed catastrophic injury leaving a durable mark. Frederick and Loewenstein's 1999 review chapter then catalogued domains of strong adaptation alongside domains of weak or absent adaptation, cautioning that adaptation is far from universal.
By the mid-2000s the pure set-point model had been overturned by panel data. Diener et al. 2006 formalized five revisions: baselines are positive rather than neutral; individuals have different set-points; a person holds multiple set-points across life domains that can move independently; set-points can change in response to circumstances; and people differ in how much and how fast they adapt. Reviewing the same longitudinal evidence, Lucas 2007 concluded plainly that for some major events the set-point shifts and does not return. The treadmill, in short, has a slope. The two most recent authoritative syntheses of the field, Diener et al. 2018 in Nature Human Behaviour and the companion review Diener et al. 2018, reaffirm this settled picture: adaptation varies by domain, and full return to baseline is the exception for severe, permanent conditions rather than the rule.
The pivotal quantitative synthesis remains Luhmann et al. 2012 in the Journal of Personality and Social Psychology, a meta-analysis of 313 samples of reactions to major life events. Its key methodological move is to separate two components of well-being that adapt differently: affective well-being (the balance of positive and negative feelings) and cognitive well-being (reflective judgments of life satisfaction). The two curves diverge (feelings often recover more completely and faster than considered life-satisfaction judgments), and the trajectory depends heavily on which event occurred. The event-specific patterns below, drawn from that meta-analysis and the underlying national panels, are the ones that matter for injury litigation.
Disability and catastrophic injury is the clearest case of incomplete adaptation, and notably the literature that speaks most directly to courts. Oswald & Powdthavee 2008 — subtitled "with implications for economists and judges" — provides longitudinal evidence that people who become disabled show genuine recovery in mental well-being, but that adaptation to severe disability is incomplete; the authors use the residual, unrecovered loss to derive the compensatory damages owed for pain and suffering, explicitly to replace the ad hoc methods courts now use. A companion analysis, Powdthavee 2009, shows part of the measured "recovery" is a statistical artifact of anticipation and focusing effects rather than true habituation — meaning naive before/after comparisons can overstate how much people adapt. And Boyce & Wood 2011 finds adaptation is conditional on personality, so a population-average recovery curve will not fit every claimant. The consistent bottom line: expect real but partial recovery for moderate injury, and little to none for severe, permanent injury.
Chronic pain and chronic illness adapt weakly. McNamee & Mendolia 2014, using the Australian HILDA panel, documents a significant, persistent negative effect of chronic pain on life satisfaction, larger where pain is more severe and enduring. The mechanism explains the poor adaptation: pain is re-instantiated daily and resists the attentional habituation that drives adaptation to static circumstances, so it is among the conditions to which people adapt least. For litigation, chronic pain warrants a largely durable loss projection.
Widowhood. Lucas et al. 2003 shows a large drop in life satisfaction at a spouse's death followed by slow, gradual recovery toward baseline (though frequently not fully back) over several years. Bereavement is thus strong initial impact with substantial but slow, often incomplete, adaptation.
Divorce and separation. Lucas 2005, "Time Does Not Heal All Wounds," finds decline that often begins before the legal event and only modest recovery afterward, with average satisfaction settling below the pre-event level.
Unemployment is the standout case of failed adaptation. Lucas et al. 2004 found that people do not return to baseline even after regaining work — a permanent downward shift, or "scar." Clark et al. 2008, testing adaptation across events in the German panel, found near-complete adaptation to marriage, divorce, widowhood and childbirth, but incomplete adaptation to unemployment. This dissociation is the crux: adaptation is not a general trait but is governed by the nature of the event. Clark & Georgellis 2013 reports a similar pattern in the British panel, and Clark et al. 2016 shows the same failure to adapt to sustained poverty — economic deprivation, like joblessness, scars.
A newer preregistered meta-analysis, Bühler et al. 2023 (44 studies, 121,187 participants), adds a calibrating note: across life events the average effects on personality, life satisfaction and self-esteem are reliable but small and event-specific, largest in the work domain. This does not contradict the litigation-relevant findings — it warns against assuming any single large universal effect and reinforces that magnitude must be read off the specific event.
The affective/cognitive split has direct forensic force. Affective adaptation — partial recovery of moment-to-moment mood, is common; people stop feeling acute distress every hour. But evaluative adaptation — recovery of the overall judgment that one's life is going well, is frequently absent for serious, permanent losses. A person with severe disability may laugh at a joke yet, rating life as a whole, register a durable, honest deficit tied to lost capabilities. Damages for "loss of enjoyment of life" map onto the evaluative measure, the one that adapts least, so judging suffering solely by whether a plaintiff can still enjoy pleasant moments systematically understates the loss.
Whether adaptation occurs is predicted by consistent moderators: severity (larger shocks recover less proportionally), permanence and recurrence (irreversible or continuously re-experienced conditions adapt poorly), controllability (uncontrollable, involuntary losses scar more than chosen transitions), and social and economic resources (which speed and complete recovery).
The science supports a defensible middle path between two errors. Assuming permanent full loss overstates damages for events with genuine adaptation — projecting the initial post-injury deficit flat across the lifespan ignores real recovery for moderate injury, grief and transitional losses. Assuming full recovery understates damages, and is the error embedded in a naive "treadmill" defense: the evidence flatly contradicts complete adaptation for chronic pain, severe disability and unemployment.
The evidentiary posture is therefore to project a partial, event-specific recovery trajectory, parameterized by three quantities the literature supplies: the initial drop, the rate/half-life of any recovery (fast for marriage and mild conditions; slow and multi-year for bereavement; effectively absent for severe disability, chronic pain and unemployment), and the long-run residual that persists. This replaces the arbitrariness of medical-bill multipliers with a transparent model: quantify the well-being deficit, apply the empirically observed adaptation curve for that injury type, integrate the residual over the expected lifespan, and monetize it. The valuation step now has its own calibrated evidence: Huang et al. 2018, using an Australian panel of 28,347 people, estimates willingness to pay of roughly A$42,000–67,000 per quality-adjusted life-year and about A$2,000 per year to avoid a long-term health condition, derived from life-satisfaction data rather than stated preference.
The honest headline for a court is neither "people get over it" nor "the loss is forever," but: it depends on the injury, the panel evidence tells us how, and for the injuries that dominate litigation (chronic pain, catastrophic disability, and their economic sequelae), adaptation is weak and the loss is largely permanent.