The full experiment involves participants coming into the lab on five separate occasions for neuropsychological testing, a decision making battery, functional and structural MRI, and two TMS sessions for stimulation of the target or control stimulation site. The clinical trial component concerns only the last two sessions where subjects will be randomly assigned to different groups to receive different TMS interventions.

In particular, the TMS experiments will ask two main questions:

1. What is the causal role of frontal pole in explore-exploit behavior in younger and older adults?
2. What is the causal role of IFG in explore-exploit behavior in younger and older adults?

The investigators will use continuous theta-burst transcranial magnetic stimulation (cTBS, Huang et al. 2005) to inhibit neural activity in each region for approximately 50 minutes (Wischnewski & Schutter, 2015) and measure the downstream effects on behavior in younger and older adults. Consistent with their respective roles in the explore-exploit circuit (Figure 5 in Research Strategy), the investigators predict that inhibition of frontal pole will lead to a selective reduction in directed, but not random, exploration, while inhibition of IFG will decrease exploitation and lead to increases in both types of exploration.

Participants in each age group will be pseudo-randomly assigned to either the frontal pole group or IFG group such that the study will have 42 participants (21 males, 21 females) in each group. Thus there will be four distinct groups of subjects older frontal pole, younger frontal pole, older IFG, younger IFG. Each participant will take part in two TMS sessions, one target and one control session. The order of sessions will be counterbalanced across subjects.

The primary endpoints of the study are to determine whether:

1. cTBS applied to frontal pole inhibits directed exploration within the younger and older groups
2. cTBS applied to IFG promotes both directed and random exploration within the younger and older groups

The study is powered to answer these questions with 80% power at a threshold of p < 0.05.

Participants will take part in five experimental sessions. These will involve neuropsychological testing, a decision making battery, functional and structural MRI, and two sessions of transcranial magnetic stimulation applied to either a target or control region. The clinical trial component concerns only the last two sessions where subjects will be randomly assigned to different groups to receive different TMS interventions. For completeness, all sessions are described below:

240 participants (110 younger, 130 older, 50% female) will be recruited from the local Tucson population. Neuropsychological testing in the first session will be used to exclude participants who exhibit symptoms of mild cognitive impairment or early stage Alzheimer's.

Decision making battery: All participants will perform the Horizon Task (Wilson et al. 2014) to measure explore-exploit behavior in addition to a Decision Making battery to measure risk taking, ambiguity attitude, temporal discounting, and behavioral variability.

Neuroimaging: 92 younger and 92 older adults will return to take part in the MRI experiment. All participants will be screened during the telephone interview and the behavioral session to rule out contraindications for MRI and TMS and screened once more at the start of the MRI session. MRI experiments will take place in the 3T Skyra scanner at the neuroimaging facility at Banner University Medical Center. All neuroimaging experiments will be run according to standard safety guidelines (Rossi et al. 2009) and participants will be screened and give informed consent before the procedure beings.

Each scan session will last about 90 minutes and will include a short localizer scan (approximately 1 minute), a T1-weighted structural scan (6 minutes), diffusion tensor imaging to assess structural connectivity (DTI, 10 minutes), FLAIR to quantify white matter hyperintensities (2 minutes), a field map to correct for magnetic field inhomogeneities, especially in frontal pole (2 minutes), resting-state fMRI to measure functional connectivity (8 minutes), and event-based fMRI using the task described below (50-60 minutes). All functional scans will be acquired with parameters that are optimized prevent signal dropout in frontal pole and other frontal regions (Deichmann et al. 2003).

Sessions 4 and 5 (TMS): The TMS sessions fit the definition of a clinical trial because subjects will be prospectively assigned to different groups who will receive different interventions in different orders. Participants in each age group will be pseudo-randomly assigned to groups according to whether the target region for TMS is frontal pole of IFG. Thus, there will be four distinct groups of subjects: older frontal pole, younger frontal pole, older IFG, younger IFG, with 42 participants (21 males, 21 females) in each group. Each participant will take part in two TMS sessions, one target and one control session. The order of sessions will be counterbalanced across subjects.

TMS sessions will take place in new TMS facility at Banner University Medical Center. All TMS will be delivered in line with established safety guidelines (Rossi et al, 2009) under the supervision of co-I Chou who is the director of the TMS facility and has extensive experience with TMS in young, old and cognitively impaired subjects. When followed correctly, these guidelines ensure that TMS is a very safe procedure with only a handful of adverse events occurring in approximately 20 years of research.

After completing consent and screening procedures, each TMS session will begin with the measurement of the action motor threshold (AMT). The AMT is defined as the minimum stimulus intensity that produces a liminal motor-evoked potential (about 200 μV in 50% of 10 trials) during isometric contraction of the tested muscle. After the AMT has been measured, participants will receive cTBS (50Hz stimulation at 80% AMT for 40 seconds; Huang et al 2005). Stimulation will be applied to either the target area (frontal pole or IFG) or control area (vertex) with the order of sessions counterbalanced across participants. Proper targeting of each region will be achieved using a frameless neuronavigation system (Localite TMS Navigator) with a Polaris Spectra infrared camera (Northern Digital, Waterloo, Ontario, Canada). This system enables stimulation to be centered on specific coordinates in Montreal Neurological Institute (MNI) space. Coordinates [x,y,z] = [35,50,15] will be used for frontal pole (based previous fMRI work and our original TMS study, Daw et al., 2006; Zajkowski et al. 2017) and [x,y,z] = [56,16,22] for right IFG based on the location of our activations in the young pilot group. The vertex control site is defined as the Cz position of a 10-20 EEG system. After receiving TMS, participants will stare at a white wall for 1 minute before performing the Horizon Task for 45 minutes.

Inclusion Criteria:

• Male or female aged 18-30 (younger adults) or 65-74 (older adults)
• No subjective memory complaints
• Fluent in English or formal education in English starting from at least the age of 5
• Telephone Interview for Cognitive Status (TICS) score > 31
• Montreal Cognitive Assessment (MoCA) > 25
• Score ≤ 10 on the Hamilton Depression Rating Scale (Hamilton, 1960)
• No significant neurological, psychiatric, medical illness or injury that would affect cognitive function
• No history of concussion with greater than 5 minutes of loss of consciousness
• No history of a psychoactive substance use disorder
• Able and willing to provide informed consent

Exclusion Criteria:

• Subjective memory complaints
• Telephone Interview for Cognitive Status (TICS) score less than or equal to 31
• Montreal Cognitive Assessment (MoCA) less than or equal to 25
• Score less than one standard deviation below the age, education, and sex adjusted mean from the NACC UDS version 3 normative cohort (Albert et al. 2011; McKhann et al., 2011
• Criteria for DSM-5 diagnosis or history of serious psychiatric disease or diagnosed learning disabilities
• Any other neurological, psychiatric, or medical illness or injury expected to interfere with cognitive function or memory including but not limited to stroke (diagnosed with evidence of stroke), head injury, epilepsy, Parkinson's, brain cancer, depression. Migraines OK. May have TIAs with no sign of impairment and no sequelae following the event
• Active substance abuse disorder i.e. alcohol, nicotine. Previous substance abuse of cocaine, Ecstasy, LSD, IV drugs
• History of seizure disorder as child or currently experiencing or on medications for seizures. Exception is febrile seizures as a child.
• Currently taking the following medications, which are contraindications for TMS: tricyclic antidepressants (Amitriptyline, Clomipramine, Doxepine, Imapramine, Maprotiline, Nortriptyline), anti-psychotic medication (Clozapine), Anti-virals (Foscarnet, Ganciclovir, Ritonavir), Bronchodilator (Theophylline), Amphetamines, gamma-Hydroxybutyrate, Ketamine
• Have recently stopped taking the following medications, which are contraindications for TMS: Alcohol, Benzodiazepines, Barbiturates, Chloral Hydrate, Meprobamate
• Any condition which may prevent the subject from adhering to the study protocol, as determined by the PI, i.e. reported learning disability, cataracts impairing vision, colorblindness.
• The presence of any metallic implant or foreign body, including dental bridges excludes participants from MRI. Removable body piercings/implants okay. Movement disorders that prevent the subject from being still for the MRI. Other contraindications to MRI including being a professional metal worker or welder, having recurring panic attacks or being claustrophobic, being pregnant, or an abnormally high weight or height to fit in scanner. Patients with these MRI contraindications will not be enrolled.

All behavioral and neuroimaging data will be shared after publication of the findings journal articles. Behavioral data from all experiments, including behavior from in the scanner will be shared on the dataverse.org. Neuroimaging data will be shared on openneuro.org an NIH sponsored repository for neuroimaging data including complex fMRI data sets. All code pertaining to data analyses used for any publications will be shared on GitHub.

All data will be deidentified before sharing.

Data will be deposited into the repositories as soon as possible but no later than within one year of the completion of the funded project period for the parent award or upon acceptance of the data for publication, or public disclosure of a submitted patent application, whichever is earlier.

Any user with an account at dataverse.org, openneuro.org or GitHub will be able to access data and/or code.

Because stimulation occurs at different locations on the scalp participants will be able to distinguish between the two conditions. However, because they will be unaware of our hypothesis, they will not know which condition is the control condition and which is the target condition.