Knowing When to Pass: The Effect of AI Reliability in Risky Decision Contexts.

OBJECTIVE: This study manipulates the presence and reliability of AI recommendations for risky decisions to measure the effect on task performance, behavioral consequences of trust, and deviation from a probability matching collaborative decision-making model. BACKGROUND: Although AI decision support improves performance, people tend to underutilize AI recommendations, particularly when outcomes are uncertain. As AI reliability increases, task performance improves, largely due to higher rates of compliance (following action recommendations) and reliance (following no-action recommendations). METHODS: In a between-subject design, participants were assigned to a high reliability AI, low reliability AI, or a control condition. Participants decided whether to bet that their team would win in a series of basketball games tying compensation to performance. We evaluated task performance (in accuracy and signal detection terms) and the behavioral consequences of trust (via compliance and reliance). RESULTS: AI recommendations improved task performance, had limited impact on risk-taking behavior, and were under-valued by participants. Accuracy, sensitivity (d'), and reliance increased in the high reliability AI condition, but there was no effect on response bias (c) or compliance. Participant behavior was only consistent with a probability matching model for compliance in the low reliability condition. CONCLUSION: In a pay-off structure that incentivized risk-taking, the primary value of the AI recommendations was in determining when to perform no action (i.e., pass on bets). APPLICATION: In risky contexts, designers need to consider whether action or no-action recommendations will be more influential to design appropriate interventions.

The effect of ageing and indomethacin on forearm reactive hyperaemia in healthy adults.

We have previously shown that non-selective cyclo-oxygenase inhibition, via indomethacin, unfavourably increased central blood pressure in older adults, with little effect in young adults. In addition, the vasoactive prostaglandins have been shown to contribute to both peripheral vasodilator responses and large artery function; however, there is little information available in older adults and conflicting reports in young adults on the extent to which resistance vessel function is influenced by indomethacin. Thus, we tested the hypothesis that cyclo-oxygenase inhibition using indomethacin would attenuate forearm vascular conductance during reactive hyperaemia in older adults compared with young adults. Forearm blood flow responses to 5 min of forearm ischaemia were measured in 26 healthy adults (13 young, 25 ± 5 years old; and 13 older, 65 ± 6 years old), using venous occlusion plethysmography before and after indomethacin. Baseline forearm blood flow and vascular conductance were not different between groups during either trial, and there were no age-related differences prior to cyclo-oxygenase inhibition. Peak forearm vascular conductance and blood flow were similar between groups before indomethacin, but lower in older adults after indomethacin compared with young adults (27 ± 4 versus 41 ± 4 ml (100 ml)(-1) min(-1) (100 mmHg)(-1), P = 0.02; and 23 ± 3 versus 33 ± 3 ml (100 ml)(-1) min(-1), P = 0.02, respectively). These results, in conjunction with our previous findings in large arteries, suggest that ageing alters the effect of cyclo-oxygenase inhibition on vascular responses, and specifically, the resistance vessel responses underlying reactive hyperaemia.

Relationship of muscle sympathetic nerve activity to insulin sensitivity.

PURPOSE: An association between insulin resistance and activation of the sympathetic nervous system has been reported in previous studies. However, potential interactions between insulin sensitivity and sympathetic neural mechanisms in healthy people remain poorly understood. We conducted a study to determine the relationship between sympathetic activity and insulin resistance in young, healthy humans. METHODS: Thirty-seven healthy adults (18-35 years, BMI <28 kg m(-2)) were studied. Resting muscle sympathetic nerve activity (MSNA) was measured with microneurography and insulin sensitivity of glucose and free fatty acid metabolism was measured during a hyperinsulinemic-euglycemic clamp with two levels of insulin. RESULTS: During lower doses of insulin, we found a small association between lower insulin sensitivity and higher MSNA (P < 0.05) but age was a cofactor in this relationship. Overall, we found no difference in insulin sensitivity between groups of low and high MSNA, but when women were analyzed separately, insulin sensitivity was lower in the high MSNA group compared with the low MSNA group of women. CONCLUSIONS: These data suggest that MSNA and insulin sensitivity are only weakly associated with young healthy individuals and that age and sex may be important modifiers of this relationship.

Sympathetic support of energy expenditure and sympathetic nervous system activity after gastric bypass surgery.

OBJECTIVE: This study was designed to determine how gastric bypass affects the sympathetically-mediated component of resting energy expenditure (REE) and muscle sympathetic nerve activity (MSNA). DESIGN AND METHODS: We measured REE before and after beta-blockade in seventeen female subjects approximately three years post-gastric bypass surgery and in nineteen female obese individuals for comparison. We also measured MSNA in a subset of these subjects. RESULTS: The gastric bypass subjects had no change in REE after systemic beta-blockade, reflecting a lack of sympathetic support of REE, in contrast to obese subjects where REE was reduced by beta-blockade by approximately 5% (P < 0.05). The gastric bypass subjects, while still overweight (BMI = 29.3 vs 38.0 kg·m(-2) for obese subjects, P < 0.05), also had significantly lower MSNA compared to obese subjects (10.9 ± 2.3 vs. 21.9 ± 4.1 bursts·min(-1) , P < 0.05). The reasons for low MSNA and a lack of sympathetically mediated support of REE after gastric bypass are likely multifactorial and may be related to changes in insulin sensitivity, body composition, and leptin, among other factors. CONCLUSIONS: These findings may have important consequences for the maintenance of weight loss after gastric bypass. Longitudinal studies are needed to further explore the changes in sympathetic support of REE and if changes in MSNA or tissue responsiveness are related to the sympathetic support of REE.

Cyclooxygenase inhibition augments central blood pressure and aortic wave reflection in aging humans.

The augmentation index and central blood pressure increase with normal aging. Recently, cyclooxygenase (COX) inhibitors, commonly used for the treatment of pain, have been associated with transient increases in the risk of cardiovascular events. We examined the effects of the COX inhibitor indomethacin (Indo) on central arterial hemodynamics and wave reflection characteristics in young and old healthy adults. High-fidelity radial arterial pressure waveforms were measured noninvasively by applanation tonometry before (control) and after Indo treatment in young (25 ± 5 yr, 7 men and 6 women) and old (64 ± 6 yr, 5 men and 6 women) subjects. Aortic systolic (control: 115 ± 3 mmHg vs. Indo: 125 ± 5 mmHg, P < 0.05) and diastolic (control: 74 ± 2 mmHg vs. Indo: 79 ± 3 mmHg, P < 0.05) pressures were elevated after Indo treatment in older subjects, whereas only diastolic pressure was elevated in young subjects (control: 71 ± 2 mmHg vs. Indo: 76 ± 1 mmHg, P < 0.05). Mean arterial pressure increased in both young and old adults after Indo treatment (P < 0.05). The aortic augmentation index and augmented pressure were elevated after Indo treatment in older subjects (control: 30 ± 5% vs. Indo 36 ± 6% and control 12 ± 1 mmHg vs. Indo: 18 ± 2 mmHg, respectively, P < 0.05), whereas pulse pressure amplification decreased (change: 8 ± 3%, P < 0.05). In addition, older subjects had a 61 ± 11% increase in wasted left ventricular energy after Indo treatment (P < 0.05). In contrast, young subjects showed no significant changes in any of the variables of interest. Taken together, these results demonstrate that COX inhibition with Indo unfavorably increases central wave reflection and augments aortic pressure in old but not young subjects. Our results suggest that aging individuals have a limited ability to compensate for the acute hemodynamic changes caused by systemic COX inhibition.