Interindividual variation in insulin sensitivity
Conditions
Interventions
Group 1: Intranasal insulin (behavioral project).
Each participant receives either 40 units of intranasal insulin (I.U.
Huminsulin Normal KwikPen 100 I.U./ml
Lilly Germany GmbH), or 0.4 ml of placebo (Saline KwikPen, Lilly Germany GmbH) on two different days in a counterbalanced order. Insulin and placebo are administered with a precision air pump (Aero Pu
Sponsors
Max-Planck-Institut für Stoffwechelforschung
Eligibility
Sex/Gender
All
Age
18 Years to 50 Years
Inclusion criteria
Inclusion criteria: - No presence or history of neurological or psychiatric disorders, or substance abuse - No severe metabolic disorders or manifest Diabetes Mellitus - No Diabetes medication or medication affecting the body weight - No medication affecting the immune system - No psychotropic medication in the last 3 months - No weight change > 5% in the last 3 months - No general contraindications with respect to MRI measurements (e.g., metal implants) - Normal or corrected-to-normal vision - BMI > 16,5 kg/m2 - Good German language skills - Women: not pregnant or nursing
Exclusion criteria
Exclusion criteria: - Reported sleep periods or excessive motion during fMRI (> 3% volumes with framewise displacement > 1 mm) - Missing responses in the Reinforcement Learning Task > 10% - Outlier with respect to asymmetry in reinforcement learning in the control condition ( 13, or item 9 score > 0 - Sickness or medication affecting the brain or the immune system - HOMA-IR > 2 (mean of both testing days) in the insulin-sensitive sample - HOMA-IR < 2 (mean of both testing days) in the sample with low insulin sensitivity - Drop out (not completing both testing days)
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| The primary outcome is reinforcement learning behavior. Reinforcement learning behavior is assessed by a computer experiment, in which the participant repeatedly chooses between two options via a key press, and then receives feedback about obtaining a reward or not. The experiment consists of 2 to 4 rounds, where each round includes 4 option pairs with different reward probabilities, presented in pseudorandom order in such a way that each pair is shown 24 times. The reinforcement learning experiment is used to assess how decision making is optimized in order to maximize the reward based on previous experiences. In Arm 1 (intranasal insulin, behavioral project), the computer experiment consists of 4 rounds each lasting 7.6 minutes, with a total duration across all 4 round of 30.4 minutes. The first round starts 10-25 minutes before the intervention, and the remaining rounds 2-4 within an interval of 10 to 80 minutes after the intervention. With this schedule, we aim to explore the temporal distribution of the effects of intranasal insulin. In Arm 2 (oral intake of glucose solution), the computer experiment consists of 2 successive rounds with a total duration of 19.7 minutes. The experiment starts 25 minutes after the intervention. In Arm 3 (intranasal insulin, fMRI project), the computer experiment also consists of 2 successive rounds with a total duration of 19.7 minutes. The start relative to intervention will be determined dependent on the results of Arm 1 (within an interval of 10 to 80 minutes after the intervention). | — |
Secondary
| Measure | Time frame |
|---|---|
| The secondary outcome in Arm 2 and 3 is the neural encoding of reward prediction errors. Within the framework of probabilistic reinforcement learning, a reward prediction error corresponds to the difference between the expected and the actual outcome of a decision. The neural encoding of reward prediction errors is assessed using functional magnetic resonance imaging (fMRI). To this end, the computer experiment is conducted in the MR scanner enabling a continuous measurement of the whole brain activity during the task. Thus, the timing of the fMRI measurement corresponds to the timing of the computer experiment described above. The neural encoding of reward prediction errors can then be computed as the correlation between i) the activity in dopaminergic brain regions ventral tegmental area and nucleus accumbens during the feedback, and ii) the size of the corresponding reward prediction errors. The fMRI-measurement during the reinforcement learning experiment takes 19.7 min and includes 1,699 volumes. In Arm 2 and 3, an additional fMRI measurement is scheduled before the computer experiment. It assesses the average brain activity before, during, and after the intervention. This additional fMRI measurement starts 2 minutes before the intervention and takes 15-30 minutes. Imaging data are acquired using a 3T MRI scanner with a 64-channel head coil (MAGNETOM Prisma Fit, Siemens AG, Medical Solutions, Erlangen, Germany). | — |
Countries
Germany
Contacts
Public ContactMarc Tittgemeyer
Max-Planck-Institut für Stoffwechselforschung
Outcome results
None listed