The neural basis of prosocial effort-based decision-making in older adults at risk for Alzheimer's disease

This study demonstrates that while older adults at risk for Alzheimer's disease (particularly APOE4 carriers) show reduced overall motivation, prosocial incentives effectively sustain effortful behavior and vigor by engaging specific frontostriatal neural circuits, suggesting a promising strategy to support goal-directed behavior in at-risk aging.

The Gist

The Big Picture: The "Charity vs. Self" Test for the Aging Brain

Imagine your brain is a smart energy manager. Every day, it has to decide: "Is this task worth the energy I have to spend to do it?"

Usually, as we get older, our energy manager gets a bit stingy. We might think, "Why walk to the store when I can order online? It's too much effort." This is called effort discounting—we start avoiding things that require hard work.

However, this study asked a fascinating question: Does this stinginess change if the reward isn't for us, but for someone else?

The researchers took a group of older adults (some of whom have a genetic risk for Alzheimer's disease) and gave them a digital game. They had to choose whether to press a keyboard button many times to earn money.

• Scenario A: The money goes into their own pocket.
• Scenario B: The money goes to a charity of their choice.

The Main Findings: The "Good Deed" Boost

1. We are more willing to work for others than for ourselves.
Even though the older adults generally preferred to save their energy, they were much more willing to press those buttons when the reward was for charity.

• The Analogy: Think of your willpower like a battery. When you are asked to do something for yourself, the battery drains fast. But when you are asked to do it for a good cause (like helping a neighbor), it's like someone just plugged in a solar charger. You suddenly have more energy to give.

2. The "APOE4" Factor: The Genetic Glitch.
Some participants carried a gene called APOE4, which is a known risk factor for Alzheimer's.

• The Glitch: These participants were generally more stingy with their energy. They were quicker to say "No" to the button-pressing task, regardless of whether the money was for them or for charity.
• The Twist: However, once they did decide to do the task for charity, they pressed the buttons faster and harder (more "vigor") than anyone else.
• The Analogy: Imagine a car with a slightly faulty engine (the APOE4 gene). It's harder to get the car started (they say "No" more often). But once you get it moving on a "good cause" road, the engine roars to life and drives with surprising speed.

What Happened Inside the Brain?

The researchers used brain scans (fMRI) to look at the "wiring" of the brain while the participants rested. They found two main things:

1. The "Social Value" Hub (vmPFC and ACC)
The brain has a specific control center (the ventromedial prefrontal cortex) that acts like a value calculator. It weighs how much a reward is worth against how hard the work is.

• The Finding: People who were more willing to work for charity had stronger connections between this "value calculator" and other parts of the brain involved in thinking about others (mentalizing).
• The Analogy: It's like having a high-speed internet cable connecting your "money calculator" to your "empathy center." When you think about helping others, the signal travels instantly, making the hard work feel worth it.

2. The "Motivation Wire" Break (in APOE4 carriers)
The people with the Alzheimer's risk gene had a weaker connection between two specific brain areas: the Nucleus Accumbens (the "go" button) and the Anterior Cingulate Cortex (the "effort monitor").

• The Analogy: Imagine a rubber band connecting the "Go" signal to the "Work" signal. In healthy brains, this band is tight and snaps back quickly to motivate action. In the APOE4 carriers, this rubber band is loose and stretched out. It takes a lot more convincing (a bigger reward) to get them to start, but once they are in "charity mode," the band snaps tight again.

Why Does This Matter?

This study offers a hopeful message for aging and Alzheimer's prevention:

• The "Prosocial" Strategy: If you want to motivate an older adult (especially one at risk for Alzheimer's) to exercise, study, or do chores, framing it as a helping behavior might work better than framing it as "self-improvement."
• The "Spark" is Still There: Even though the APOE4 gene makes the brain more resistant to effort, the "spark" for doing good things remains strong. In fact, it might be the most powerful motivator available.

The Bottom Line

As we age, our brains naturally get a bit lazy about working for ourselves. But our brains are wired to love helping others. Even in brains at risk for Alzheimer's, the desire to do good can override the fatigue, lighting up the neural pathways that keep us active and engaged.

In short: If you want to get an older adult moving, don't just tell them, "Do this for your health." Tell them, "Do this to help someone else." Their brain will thank you.

Technical Summary

1. Problem Statement

Alzheimer's disease (AD) and aging are associated with impairments in effort-based decision making (the process of weighing reward value against effort costs), which undermines autonomy and health behaviors. While older adults typically show increased aversion to effort, aging is also characterized by a shift toward socioemotional and prosocial goals (Socioemotional Selectivity Theory).

• The Gap: It is unclear how prosocial incentives (benefiting others) versus self-oriented incentives affect effort-based decision making in cognitively unimpaired older adults at genetic risk for AD.
• The Genetic Factor: The APOE4 allele is the strongest genetic risk factor for late-onset AD. Carriers often exhibit early alterations in frontostriatal reward circuits (e.g., reduced connectivity between the ventral tegmental area, nucleus accumbens, and prefrontal cortex) before cognitive decline manifests.
• Objective: To investigate whether prosocial incentives can motivate effort in at-risk older adults, identify the neural substrates (specifically resting-state functional connectivity, rsFC) supporting these decisions, and determine how APOE4 carrier status modulates these behavioral and neural processes.

2. Methodology

Participants:

• Sample: 52 cognitively unimpaired older adults (Mean age = 68.48; 38 females) from the PREVENT-AD cohort (McGill University).
• Risk Profile: All had a first-degree family history of AD.
• Genotyping: 18 were APOE4 carriers; 34 were non-carriers.
• Exclusions: 3 participants were excluded from neuroimaging due to motion artifacts.

Experimental Task:

• Effort-Based Decision-Making Task: 120 trials where participants decided whether to exert physical effort (keyboard presses) for monetary rewards.
• Conditions:
  1. Self-Oriented: Earn money for oneself.
  2. Prosocial: Donate money to a charity of choice.
• Variables: Reward amounts ($0.50–$3.00) and Effort levels (3, 20, 37, 54, or 70 key presses).
• Metrics: Acceptance/Rejection decisions, Response Time (RT), and Vigor (button presses per second during work trials).

Neuroimaging:

• Acquisition: Multi-echo resting-state fMRI (10 min 24 sec, eyes-closed) on a 3T Siemens Prisma scanner.
• Preprocessing: fMRIPrep v22.1.1, multi-echo denoising (tedana), motion correction, and normalization to MNI space.
• Analysis:
  • Seed-to-Voxel: Analyzed connectivity from ventromedial prefrontal cortex (vmPFC) and anterior cingulate cortex (ACC) seeds.
  • ROI-to-ROI: Examined connectivity within a frontostriatal reward network (VTA, NAcc, dACC, vmPFC) comparing APOE4 carriers vs. non-carriers.
  • Multivariate Pattern Analysis (fc-MVPA): Data-driven approach to identify whole-brain connectivity patterns associated with behavioral differences.

Computational Modeling:

• Modeling: Reward-effort curves were fitted using Linear, Sigmoid, and Weibull functions.
• Selection: Random-effects Bayesian model comparison determined the best-fitting model.
• Parameters: Extracted Bias (tendency to avoid effort) and Sigma (reward sensitivity) from the winning Sigmoid model.

Statistical Analysis:

• Mixed-effects models (GLMM and robust LMM) for behavioral data.
• Linear mixed-effects models for computational parameters.
• FDR correction for multiple comparisons in neuroimaging.

3. Key Contributions

1. Prosocial Motivation in Preclinical AD: Demonstrated that older adults at risk for AD are more willing to exert effort for prosocial rewards than self-oriented rewards, suggesting preserved or enhanced socioemotional motivation in early disease stages.
2. Neural Correlates of Prosocial Effort: Identified specific resting-state functional connectivity (rsFC) patterns (involving vmPFC and ACC) that predict an individual's bias toward prosocial effort.
3. APOE4-Specific Neural Disruption: Revealed that APOE4 carriers exhibit reduced connectivity in the dACC–NAcc circuit, which correlates with a general aversion to effort, yet they retain the ability to mobilize high vigor specifically for prosocial goals.
4. Computational Characterization: Used computational modeling to show that the relationship between reward and effort follows a sigmoid function, with APOE4 carriers showing a steeper "bias" against effort expenditure overall.

4. Key Results

Behavioral Findings:

• Prosocial Preference: Participants accepted significantly more effort trials for prosocial rewards compared to self-oriented rewards (OR = 1.98).
• Response Time: Decision times were significantly longer for prosocial rewards, suggesting increased deliberative processing.
• Vigor: No overall difference in vigor between conditions for the whole group.
• APOE4 Effects:
  • Carriers showed a higher overall bias away from effort (lower acceptance) compared to non-carriers across both conditions.
  • Interaction: Carriers displayed significantly higher vigor (faster/more forceful pressing) specifically when exerting effort for prosocial rewards compared to self-oriented rewards. Non-carriers showed no such difference.

Computational Modeling:

• The Sigmoid function was the best fit for the data (Protected Exceedance Probability > 0.999).
• Bias Parameter: Significantly higher (more effort-averse) in the self-oriented condition compared to prosocial.
• APOE4 Interaction: Carriers had a significantly higher bias parameter (steeper discounting of effort) than non-carriers.

Neural Findings (rsFC):

• Prosocial Bias: Higher bias toward prosocial effort (vs. self) was associated with:
  • Enhanced connectivity: vmPFC ↔ dorsomedial PFC (dmPFC); ACC ↔ left inferior frontal gyrus (IFG).
  • Reduced connectivity (Anticorrelation): ACC ↔ lateral occipital cortex, superior parietal lobule, and dorsolateral PFC (dlPFC). This suggests a shift from external attentional processing to internal socioemotional valuation.
• APOE4 Effects:
  • Carriers had reduced rsFC between the right Nucleus Accumbens (NAcc) and dorsal ACC (dACC) compared to non-carriers.
  • This reduced NAcc–dACC connectivity was positively associated with a greater bias away from effort expenditure.
  • Note: fc-MVPA did not identify distinct whole-brain patterns differentiating carriers from non-carriers, highlighting the specificity of the NAcc–dACC circuit.

5. Significance and Implications

• Preserved Prosocial Circuitry: Despite genetic risk and general effort aversion, the neural circuitry supporting prosocial motivation (vmPFC/ACC connectivity) remains functional in APOE4 carriers.
• Therapeutic Strategy: Prosocial incentives may be a potent, underutilized strategy to motivate effortful behaviors (e.g., physical activity, cognitive training) in older adults at risk for AD. Even though APOE4 carriers are generally more effort-averse, they can be "energized" (high vigor) by prosocial goals.
• Early Biomarkers: The reduced dACC–NAcc connectivity in APOE4 carriers serves as a potential early biomarker for motivational deficits that precede overt cognitive decline (MCI/AD).
• Mechanism of Action: The findings suggest that APOE4 carriers may have diminished tonic dopaminergic signaling (lower baseline motivation) but preserved or enhanced phasic responses to salient, socially meaningful rewards.

Limitations:

• Sample was homogeneous (mostly female, highly educated, Caucasian, from Quebec).
• Monetary rewards were used; real-world prosocial acts are often intrinsically rewarding.
• Cross-sectional design limits causal inferences regarding disease progression.

Conclusion:
Leveraging prosocial motivation targets a neural circuit (vmPFC/ACC) that appears resilient in preclinical AD. This approach could help sustain goal-directed behavior and support brain health in aging populations, particularly for those with the APOE4 risk allele who otherwise show reduced motivation.