Source-linked AI summary
A Cellular Doctrine of Morality: Intrinsic Active Precision and the Mind-Reality Overload Dilemma
Ahsan Adeel
TL;DR
The paper addresses how information flows that lack intrinsic checks of precision and validity may contribute to overload and poor judgment. It proposes a pyramidal-neuron-inspired active precision approach that gates signals through contextual coherence. Prior evaluations of the formulation report faster learning and lower computational cost, while the paper notes that finite resources and uncertainty remain.
Problem
The paper asks how information can be assessed for precision, validity, and relevance before attention or propagation, given that current AI systems amplify information without such intrinsic checks.
Source: §1 · pp. 1–2Method
The paper proposes Cooperative Context-sensitive Cognitive Computation (Co4), using active precision to gate signals according to agreement between internal context and external evidence.
Source: §5 · p. 5Results
Takeaways & Limitations
Coherence gating filters incoherent signals early and may reduce overload while supporting representations that are coherent, valid, and adequate to guide action.
Source: §5 · p. 5, §8 · p. 9The condition SR > SC cannot be entirely eliminated because systems have finite resources, and uncertainty is inherent in complex environments.
Source: §7 · p. 9
Abstract
from arXiv · showhide
Current AI systems, grounded in oversimplified neuroscience, risk eroding the distinction between truth and falsehood. They maximize reward by amplifying attention to information without intrinsic precision mechanisms to assess whether it is valid or worth attending to. This increases both the volume of information and the inherent biases in what the system attends to, whether true, false, or irrelevant. If not corrected, this trend will accelerate, threatening to overload systems and individuals with biased and dubious information and increasing the risk of confusion, poor judgment, and irrational or harmful decisions and behaviour, a condition I term the mind-reality overload dilemma. I argue that this threat may be mitigated by providing the public with access to more advanced AI tools built on the biophysical dynamics of pyramidal neurons underlying awake thought and higher-order cognition. These neurons support an intrinsic active precision mechanism that, rather than merely maximizing reward, uses locally and globally coherent predictions to evaluate the validity and contextual adequacy of evidence before it is attended to or propagated through hierarchies, prioritizing coherence and adequacy before attention.~While this approach does not derive or prescribe moral rules from biology, it may give rise to AI with more "real understanding", helping restore epistemic conditions by reducing information overload and amplifying reliable information, thereby supporting the formation of better-informed beliefs and more coherent judgments that benefit society at large-though no guarantees exist.
1. Mind–Reality Overload Dilemma
The mind–reality overload dilemma arises when incoming information outpaces the mind’s ability to interpret it coherently, leaving attention vulnerable to unreliable or irrelevant signals. The paper argues that coherence-based active precision could improve information selection without deriving moral rules from biology.
1. Mind–Reality Overload Dilemma
- The dilemma occurs when incoming information exceeds the brain’s capacity to assign reliable precision and establish validity, increasing confusion, misjudgment, and harmful decisions.The paper frames this as observable reality arriving faster than coherent cognition can keep up.Source: §1 · p. 1
- The overload ratio compares the speed of observable information with the speed at which the mind can segregate and attend to relevant signals.SR is the rate at which relevant and irrelevant information reaches the individual; SC is the rate of coherent cognition.Source: §1 · pp. 1–2
- Local and global coherence across partially independent sources can help estimate precision and relevance, but coherence does not guarantee truth or action-adequate representations.Adequacy means context-dependent sufficiency to guide attention and action reliably.Source: §1 · p. 1
- Current AI systems amplify attention without intrinsic checks of precision, validity, or relevance, while active precision could select coherent, valid, action-adequate signals before propagation.The proposed approach aims to improve epistemic conditions, not derive moral rules from biology.Source: §1 · p. 2
2. Origins of Mind–Reality Overload Dilemma
The section traces overload partly to simplified neural models that collapse sensory evidence and internal context, and to predictive-processing frameworks that assess precision after comparison rather than before propagation.
2. Origins of Mind–Reality Overload Dilemma
- Simplified integrate-and-fire point neurons collapse contextual predictions and sensory evidence into one feedforward scalar, limiting contextual selection and filtering.The paper also argues that attention-only transformer representations are formed without intrinsic validation against internal context.Source: §2 · p. 3
- Predictive coding and active inference compare sensory signals with internal predictions, then use precision to weight prediction errors.Precision is estimated reliability; it may be fixed or learned as a state-dependent gain.Source: §2 · p. 3
- Because these frameworks do not explicitly assess coherence, precision, or adequacy before error computation, many signals compete for limited capacity and may impede coherent interpretation.Misleading early inferences can require multiple cycles to correct or fail to converge under limited resources [32] [33].Source: §2 · p. 3
3. AI and Mind–Reality Misalignment
The section argues that contemporary AI allocates attention to learned representations without intrinsic, moment-by-moment coherence checks, so irrelevant or biased signals can propagate before filtering. Attention-optimized information platforms show a related pattern.
3. AI and Mind–Reality Misalignment
- Transformer-like systems compute learned Q, K, and V representations without explicit intrinsic precision or moment-by-moment prediction checks for coherence before attention is allocated.Filtering is described as delayed and emergent rather than intrinsic and immediate.Source: §3 · p. 3
- In these architectures, irrelevant or biased signals can propagate through the network before later filtering, while training refines relevance only after substantial processing.The paper notes that this can require deeper architectures and extensive training.Source: §3 · pp. 3–4
- Engagement-optimized platforms amplify content using attention signals without explicit prior coherence or reliability evaluation, increasing effective information flow relative to coherent interpretation capacity.The paper links this imbalance to increased likelihood of SR > SC and overload.Source: §3 · p. 4
4. Advances in Neuroscience and Computational Models
Cellular neurobiology and biophysical modeling motivate an intrinsic precision mechanism in pyramidal neurons, where coordinated sensory and contextual inputs can support coherence-sensitive processing. The mechanism varies across mental states and may help evaluate information before attention.
4. Advances in Neuroscience and Computational Models
- Layer 5 pyramidal two-point neurons integrate sensory evidence at basal dendrites and contextual input at apical dendrites; simultaneous depolarization produces bursts that amplify coherent signals and suppress incoherent activity.The proposed processing supports context-sensitive evaluation before attention and action-adequate representations.Source: §4 · p. 4
- Bursting probability varies with the relative strengths of sensory evidence and context across slow-wave sleep, wakefulness, and REM sleep.During awake thought, both inputs are high and bursting probability is maximal; during slow-wave sleep, context has no effect on output.Source: §4 · p. 4
- Cooperative context-sensitive computation can generate internal predictions that help evaluate incoming information, establish validity, and guide attention toward signals worthy of propagation.The paper associates progressively stronger contextual input, from slow-wave sleep through REM sleep to awake thought, with higher mental states.Source: §4 · p. 4
5. Future Machines
The proposed Co4 framework uses pyramidal-neuron-inspired active precision to assess coherence and contextual adequacy before signals compete for attention. The paper presents benchmark findings and an illustrative control simulation in which coherence-first processing is associated with more stable behavior.
5. Future Machines
- The framework defines active precision as agreement between external evidence and internal context, using locally and globally coherent predictions to validate signals before and during attention.Context can include online feedback, while bounded precision and feedback help preserve initially unlikely hypotheses that later evidence supports.Source: §5 · p. 5
- Co4 gates incoherent signals before attention competition, reducing effective signal propagation and helping maintain mind–reality synchrony while lowering overload risk.The paper identifies this approach as Cooperative Context-sensitive Cognitive Computation (Co4) [33].Source: §5 · p. 5
5.1. Experimental Framework for Testing the Overload Hypothesis
- In the resource-constrained control simulation, Co4 maintains alignment with the intended trajectory while the attention-only agent drifts, despite identical external inputs and conditions.The caption presents this behavioral contrast as consistent with overload mitigation, not definitive proof.Source: Figure 3 · p. 7, §5.1 · p. 7
5. Future Machines
- Co4 evaluations report significantly faster learning and reduced computational cost on vision and reinforcement-learning benchmarks, with ablations showing degraded performance when active-precision structure is modified [33].The paper describes these findings as presented in [33]; it does not provide benchmark values in this passage.Source: §5 · p. 7
5.1. Experimental Framework for Testing the Overload Hypothesis
- The driving contrast is illustrative; testing a causal link requires systematic evaluation under controlled variation of informational load.The paper attributes the Transformer agent’s drift to task-irrelevant signals competing for capacity and describes Co4 as attenuating signals that do not fit the evolving task context.Source: §5.1 · p. 8
6. Reasons to Have Hope For Humanity
Coherence- and adequacy-driven information systems may help reduce overload and support more informed, cooperative decisions. Their societal effects remain indirect, and no system can fully eliminate misuse.
6. Reasons to Have Hope For Humanity
- More coherent information environments may support informed, cooperative, and ethically grounded decisions, but outcomes also depend on incentives, institutions, and social dynamics.Source: §6 · p. 8
- Coherence-based filtering before attention may reduce misleading information and support representations adequate for guiding action.The proposed shift prioritizes coherence and reliability rather than amplifying signals that maximize attention.Source: §6 · p. 8
- No system can fully eliminate misuse, and the link between informational coherence and broader societal outcomes remains indirect.Source: §6 · p. 8
7. Possible Doubts, Objections, and Paradoxes
The paper responds to concerns about truth, motivation, autonomy, and over-convergence while acknowledging finite cognitive resources and uncertainty. It presents coherence as a support for judgment, not a determinant of values or behavior.
7. Possible Doubts, Objections, and Paradoxes
- Transparency and avoiding bias remain critical challenges for systems that assess coherence or validity.Source: §7 · p. 8
- Improved information alone may not guarantee better behavior because emotions, incentives, and social dynamics also shape human action.Coherence does not eliminate all sources of error or conflict, but may provide a more stable basis for informed decisions.Source: §7 · p. 8
- Coherence-driven systems aim to reduce exposure to misleading information rather than prescribe decisions, but whether this guidance preserves meaningful autonomy remains open.Source: §7 · pp. 8–9
- Coherence need not produce uniformity: people in different internal states may form divergent but internally consistent interpretations.The paper suggests disagreement could become more grounded and interpretable rather than driven by misinformation.Source: §7 · p. 9
- Finite resources mean the condition SR > SC cannot be eliminated, and uncertainty remains inherent in complex environments.The stated aim is to manage uncertainty by structuring information flow, not eliminate it.Source: §7 · p. 9
- Cellular mechanisms shape information filtering and integration without encoding human values or determining the content of value-guided behavior.Source: §7 · p. 9
8. Concluding Remarks
The paper proposes a cellular doctrine of morality grounded in cooperative biophysical foundations of higher-order cognition. It may help restore conditions for reliable belief formation, but it does not guarantee moral decisions.
8. Concluding Remarks
- The proposed cellular doctrine of morality is grounded in the cooperative biophysical foundations of higher-order cognition.Source: §8 · p. 9
- These foundations may help restore conditions for distinguishing truth from falsehood and forming reliable beliefs that underlie moral judgment.Source: §8 · p. 9
- No technology can guarantee moral decision-making; whether morality is intrinsic to cellular processes underlying higher-order cognition remains an open question.Source: §8 · p. 9
A. Varying Strengths of RF and CF Across Distinct Mental States: An Illustrative Example
The section contrasts rapid intuitive processing with slower, reflective interpretation. It presents high-level perception as sufficient for routine judgments and awake thought as supporting deeper contextual interpretation.
A. Varying Strengths of RF and CF Across Distinct Mental States: An Illustrative Example
- High-level perceptual processing supports rapid, intuitive judgments and routine interactions, with RF and CF input strength ranging from moderate to high.The passage associates this state with cholinergic, noradrenergic, and orexinergic systems.Source: A. Varying Strengths of RF and CF… · p. 11
- Slower, deliberate interpretation can reveal context-dependent readings of visually identical characters.Awake thought involves high-to-maximal RF and CF input strength and interaction; “deep” refers to complex integration of diverse CFs at the cellular level, enabled by TPNs.Source: A. Varying Strengths of RF and CF… · pp. 11–12
- The riddle example combines rapid intuitive processing in high-level perception with deliberate refinement in the awake thought state.The contrast is between fast initial interpretation and slow reflective refinement.Source: Figure 4 · p. 12