Physical Autonomy and Self-Determination: Building a Need-Based Framework to Guide the Development of Neurotechnology

Abstract

Neuromodulation and brain-computer interface technologies present patients, caregivers, clinicians, engineers, and regulators with significant ethical challenges surrounding patient autonomy and self-determination. Existing frameworks address autonomy and self-determination narrowly, typically focusing on the restoration of some physical deficit lost to disease, and cannot account for the trade-offs these therapeutics impose across multiple dimensions of a patient’s life. We propose a need-based framework that reconceives autonomy as a graded configuration of self-determination consisting of two procedural capacities: decisional and executional autonomy. Each of these domains is shaped by three influence dimensions consisting of mental, relational, and physical autonomy. This structure captures how neurotechnological design and deployment can simultaneously enhance and erode self-determination. We offer a candidate formalization of our framework and apply it to a well-documented case of a Dutch Parkinson’s disease patient who accepted cognitive side effects in exchange for restored motor function and demonstrate that our framework reveals trade-offs in autonomy that existing approaches cannot capture. We further show that poorly considered neurotechnology implementation can erode mental, relational, or physical autonomy independently of decisional capacity. Finally, applied to current and emerging neurotechnology, the framework unifies disparate ethical concerns of implant abandonment, neural data governance, and closed-loop neural control, under a single criterion: net impact on patient self-determination.

Mental (m) Physical (p) Relational (r) Decisional (D) Executional (E) Self-Determination (SD)

How the framework works

Move the three influence sliders in either panel and watch the two procedural capacities — and overall self-determination — update in real time.
Influence domain
Three dimensions you can set independently: mental, physical, relational autonomy (0 → 1).
Procedural domain
A weight matrix W maps them to decisional (D) and executional (E) capacity.
Self-determination
SD combines D and E; it is highest when both are strong and balanced.
Quick comparisons:
A
.82
.82
.82
Decisional
.82
Executional
.82
Self-Det.
.82
B
.30
.85
.52
Decisional
.48
Executional
.63
Self-Det.
.54

Side-by-side: the procedural domain

Each situation becomes a vector (D, E). Its length is overall capacity; its angle θ is the balance between deciding and acting. Self-determination is largest along the 45° line and collapses toward either axis.
A · SD
.82
B · SD
.54
Balance angle
θA = 45°
Balance angle
θB = 53°
Difference in SD
–.28

The weight matrix W

W decides how much each influence dimension feeds each procedural capacity. Rows are normalized to sum to 1. Drag the weights and watch the transformation change.
.85
.62
.58
m
r
p
Decisional
.50
.30
.20
Executional
.25
.25
.50
Defaults follow the paper: mental loads most on D, physical most on E. The weights are not universal — they shift by clinical population and decision context. For example, a physical deficit from blindness loads heavily on the decisional row (it shapes the option set), whereas a lower-limb amputation loads more on the executional row.

Transformation

Influence dimensions flow through W into the procedural capacities, then converge into self-determination.
Decisional D
.71
Executional E
.64
Self-Det. SD
.95
Eq. 1 · Influence → Procedural: D = wdr·r + wdm·m + wdp·p E = wer·r + wem·m + wep·p Eq. 2 · Procedural → SD: SD = √(D² + E²) · sin 2θ,  θ = arctan(E/D) (shown normalized to 0–1)