Aims/hypothesis: Approximately 25% of people with type 1 diabetes have suppressed counterregulatory hormonal and symptomatic responses to insulin-induced hypoglycaemia, which renders them at increased risk of severe, disabling hypoglycaemia. This is called impaired awareness of hypoglycaemia (IAH), the cause of which is unknown. We recently proposed that IAH develops through habituation, a form of adaptive memory to preceding hypoglycaemia. Consistent with this hypothesis, we demonstrated restoration of defective counterregulatory hormonal responses to hypoglycaemia (referred to as dishabituation) in a rodent model of IAH following introduction of a novel stress stimulus (high intensity training [HIT]). In this proof-of-concept study we sought to further test this hypothesis by examining whether a single episode of HIT would amplify counterregulatory responses to subsequent hypoglycaemia in people with type 1 diabetes who had IAH (assessed by Gold score ≥4, modified Clarke score ≥4 or Dose Adjustment For Normal Eating (DAFNE) hypoglycaemia awareness rating 2 or 3). The primary outcome was the difference in adrenaline response to hypoglycaemia following both a single episode of HIT and rest. Methods: In this randomised, crossover study 12 participants aged between 18 and 55 years with type 1 diabetes for ≥5 years and an HbA1c <75 mmol/mol (9%) were recruited. Individuals were randomised using computer generated block randomisation to start with one episode of HIT (4 × 30 s cycle sprints [2 min recovery] at 150% of maximum wattage achieved during V̇ O 2 peak assessment) or rest (control). The following day they underwent a 90 min hyperinsulinaemic–hypoglycaemic clamp study at 2.5 mmol/l with measurement of hormonal counterregulatory response, symptom scores and cognitive testing (four-choice reaction time and digit symbol substitution test). Each intervention and subsequent clamp study was separated by at least 2 weeks. The participants and investigators were not blinded to the intervention or measurements during the study. The investigators were blinded to the primary outcome and blood analysis results. Results: All participants (six male and six female, age 19–54 years, median [IQR] duration of type 1 diabetes 24.5 [17.3–29.0] years, mean [SEM] HbA1c 56 [3.67] mmol/mol; 7.3% [0.34%]) completed the study (both interventions and two clamps). In comparison with the rest study, a single episode of HIT led to a 29% increase in the adrenaline (epinephrine) response (mean [SEM]) (2286.5 [343.1] vs 2953.8 [384.9] pmol/l); a significant increase in total symptom scores (Edinburgh Hypoglycaemia Symptom Scale: 24.25 [2.960 vs 27.5 [3.9]; p<0.05), and a significant prolongation of four-choice reaction time (591.8 [22.5] vs 659.9 [39.86] ms; p<0.01] during equivalent hypoglycaemia induced the following day. Conclusions/interpretation: These findings are consistent with the hypothesis that IAH develops in people with type 1 diabetes as a habituated response and that introduction of a novel stressor can restore, at least partially, the adapted counterregulatory hormonal, symptomatic and cognitive responses to hypoglycaemia. Trial registration: ISRCTN15236211.