When Reality Outpaces Cognition: A New Scaling Principle for Reliable Intelligence

Ahsan Adeel; Talha Bin Riaz

cs.AIOct 14, 2026

When Reality Outpaces Cognition: A New Scaling Principle for Reliable Intelligence

Authors: Ahsan Adeel, Talha Bin Riaz

Organizations: University of Stirling

Abstract

This paper challenges the understanding capabilities of existing AI paradigms and emerging world models by asking a fundamental question: How quickly can reality change before an intelligent system can no longer reliably perceive, reason, and act? We formulate this as a theory- and model-agnostic challenge in which systems must infer reliable actions for increasingly distant future states across a growing observation--reality gap. We operationalize this challenge through the effective Speed of Coherent Cognition ( $S_C^{\mathrm{eff}}$ ), defined as the highest tested Speed of Reality ( $S_R$ ) through which a system continuously satisfies a predefined behavioural reliability criterion from baseline. As a proof of concept, reinforcement-learning agents are trained at the baseline $S_R$ and evaluated, without retraining, as $S_R$ is increased by up to $50\times$ while computational resources remain fixed. We compare Transformer and gated-MLP baselines with $Co^4$ , an architecture inspired by cellular mechanisms implicated in conscious processing and awake thought. Across the tested scenarios, $Co^4$ achieves substantially higher $S_C^{\mathrm{eff}}$ than the baselines, motivating $S_C^{\mathrm{eff}}$ as a new benchmark and the scaling of coherent cognition as a principle for reliable AI.

Figures & tables

Figure 2: Operational challenge of when reality outpaces cognition. The agent is trained at the baseline Speed of Reality ( $S_{R,0}=1$ ), but tested at $50\times$ the baseline ( $S_{R}=50$ ) while computational resources remain fixed. From State 1, it must infer the action appropriate for the future State 50 while States 2–50 remain unobserved. The challenge is whether the agent has learned sufficiently generalisable task-relevant world dynamics to infer a reliable action across this observation–reality gap.

Figure 3: Training performance in the easy vision-guided quadrupedal locomotion environment. Running average reward over training epochs for the Transformer, gMLP, and two $Co^{4}$ variants at $S_{R,0}=1$ . In the easy environment, obstacles are more widely separated, allowing all architectures to learn the locomotion task, although with different learning trajectories and final performance levels.

Figure 4: Scaling $S_{C}^{\mathrm{eff}}$ in the easy vision-guided quadrupedal locomotion task. Reward is shown as $S_{R}$ increases from the training baseline $S_{R,0}=1$ to $S_{R}=50$ without retraining. The dashed line denotes the predefined reliability threshold $\tau=500$ .

$\mathbf{S_{R}}$	$\mathbf{Co^{4}}$ -AA	$\mathbf{Co^{4}}$ -ADA	Transformer	gMLP
1	636.0 $\pm$ 38.2	617.3 $\pm$ 34.2	625.2 $\pm$ 37.5	415.1 $\pm$ 22.1
2	636.4 $\pm$ 38.2	615.1 $\pm$ 36.9	623.2 $\pm$ 37.4	418.2 $\pm$ 23.6
3	636.1 $\pm$ 38.2	649.2 $\pm$ 39.0	246.3 $\pm$ 14.8	166.0 $\pm$ 10.0
5	564.9 $\pm$ 33.9	605.4 $\pm$ 36.3	243.2 $\pm$ 14.6	150.0 $\pm$ 9.0
10	519.3 $\pm$ 31.2	628.2 $\pm$ 37.7	301.2 $\pm$ 18.1	149.8 $\pm$ 9.0
20	509.0 $\pm$ 30.5	188.3 $\pm$ 11.3	35.8 $\pm$ 2.1	36.9 $\pm$ 2.2

Table 1: Performance in the easy vision-guided quadrupedal locomotion task as

S_{R}

increases. Values are mean

\pm

standard deviation.

Figure 5: Representative attention patterns in vision-guided quadrupedal locomotion. The Transformer exhibits comparatively diffuse attention (top): when the obstacle is distant, attention extends substantially over background regions, while proximity to the obstacle produces broadly distributed attention across the visual input. In contrast, Co 4 (middle) exhibits more selective attention, concentrating on the immediately relevant traversable region while the obstacle is distant and shifting toward the obstacle as it becomes locally relevant. This difference is also reflected in the sequence-level attention maps (bottom), where Co 4 produces a substantially sparser pattern.

Figure 6: Training performance in the complex vision-guided quadrupedal locomotion environment. Running average reward during training for the Transformer and gMLP baselines and Co 4 variants implementing different forms of apical modulation [ 17 , 20 ] . Performance differs substantially across modulation conditions. Disabling the Co 4 modulation equations provides a control for assessing the functional contribution of contextual modulation.

Figure 7: Scaling $S_{C}^{\mathrm{eff}}$ in the complex vision-guided quadrupedal locomotion task. Reward is shown as $S_{R}$ increases from the training baseline $S_{R,0}=1$ to $S_{R}=50$ without retraining. The dashed line denotes the predefined reliability threshold $\tau=300$ . Although the three-layer gMLP performed comparably to the two-layer $Co^{4}$ during training, it showed substantially weaker generalisation as $S_{R}$ increased. Further architectural optimization, including increasing the number of layers, is beyond the scope of the present general challenge.

$\mathbf{S_{R}}$	$\mathbf{Co^{4}}$ -ADA	$\mathbf{Co^{4}}$ -AA	Transformer	gMLP
1	503.8 $\pm$ 28.6	458.0 $\pm$ 27.5	311.2 $\pm$ 18.7	232.8 $\pm$ 14.0
2	430.4 $\pm$ 26.6	466.8 $\pm$ 28.0	304.7 $\pm$ 18.3	241.8 $\pm$ 13.8
3	468.5 $\pm$ 28.6	446.2 $\pm$ 26.8	223.2 $\pm$ 13.4	201.7 $\pm$ 12.0
5	439.2 $\pm$ 27.2	488.0 $\pm$ 29.3	227.6 $\pm$ 13.7	199.4 $\pm$ 11.1
10	479.8 $\pm$ 27.2	428.3 $\pm$ 25.7	149.7 $\pm$ 9.0	187.4 $\pm$ 8.0
20	302.0 $\pm$ 18.5	321.3 $\pm$ 19.3	118.2 $\pm$ 7.1	108.7 $\pm$ 6.8

Table 2: Performance in the complex vision-guided quadrupedal locomotion environment as

S_{R}

increases. Values are mean

\pm

standard deviation.

Figure 8: Learning curves for the car-racing task.

Figure 9: Scaling the Speed of Coherent Cognition in the car-racing task. Mean reward for the Transformer baseline and Co 4 as $S_{R}$ increases from $1$ to $7$ , without retraining and with computational resources held fixed. The dashed line indicates the predefined reliability threshold, $\tau=300$ . The Transformer remains reliable through $S_{R}=2$ and Co 4 through $S_{R}=4$ , yielding $S_{C}^{\mathrm{eff}}=2$ and $S_{C}^{\mathrm{eff}}=4$ , respectively.

$S_{R}$	Co 4	Transformer
$1$	$857.89\pm 87.52$	$896.71\pm 91.02$
$2$	$794.74\pm 78.28$	$345.90\pm 35.02$
$3$	$641.25\pm 64.77$	$126.97\pm 12.42$
$4$	$329.49\pm 32.16$	$57.34\pm 5.93$
$5$	$216.50\pm 21.11$	$61.18\pm 6.48$
$6$	$116.39\pm 11.46$	$41.45\pm 4.02$

Table 3: Performance in the car racing task as

S_{R}

increases. Values are mean

\pm

standard deviation

AI Agent Reliability

AI agent reliability studies whether tool-using systems act correctly, consistently, safely, and recoverably across changing tasks and environments.

See all 339 papers

Jun 5, 2026

Think Fast: Estimating No-CoT Task-Completion Time Horizons of Frontier AI Models

Many efforts to ensure frontier AI models are safe rely on monitoring their chain-of-thought (CoT) reasoning. If models become able to perform sufficiently complex reasoning internally, without explicit thinking tokens, this would undermine such oversight. We measure how well frontier models reason without CoT across a suite of over 30,000 questions spanning 43 benchmarks in domains including math, coding, puzzles, causality, theory-of-mind, and strategic reasoning. To compare models against humans, we estimate the

50\%

-task-completion time horizon (TH): the human time required for tasks a model completes with

50\%

success rate. We complement this with a

50\%

reasoning token horizon: the minimum number of o3-mini reasoning tokens needed for tasks a model solves with

50\%

success rate. We find that the no-CoT

50\%

TH of frontier models has been doubling roughly every year over the past six years, with GPT-5.5's TH reaching over 3 minutes and reasoning token horizon exceeding 1,500 tokens. Our median estimates predict that frontier no-CoT THs could exceed 7 minutes by 2028, and 25 minutes by 2030, though these projections carry substantial uncertainty. We recommend frontier developers track this explicitly.

Dewi Gould, Francis Rhys Ward, Anders Cairns Woodruff +18

Redwood Research · Astra Fellows Program · Aether Research +10

LLM Evaluation AI Safety Evaluation LLM Reasoning Large Reasoning Models

Ask PDF

Jun 1, 2026

Unveiling the Entropy Dynamics of Chain-of-Thought Reasoning

This paper investigates the entropy dynamics of Chain-of-Thought (CoT) and uncovers a consistent two-phase structure: an Uncertainty Region of exploration transitioning sharply to a Confidence Region of convergence. We demonstrate that the Confidence Region possesses two critical properties: 1) High Reliability -- answers in the confidence region become highly accurate and stable, and 2) High Redundancy -- models generate unnecessary tokens long after reaching the correct answer. These properties unlock more efficient and reliable inference strategies: 1) Early Exit leverages reliability and redundancy to terminate computation safely when returns diminish, and 2)Test-Time Scaling uses the Confidence Region signal to prioritize converged trajectories. To operationalize these insights, we formulate Confidence Region detection as a sequential change-point detection problem, being the first to apply classical change-point methods to monitor CoT reasoning. Using the Cumulative Sum (CUSUM) algorithm, a statistically optimal change-point detector, we develop a training-free framework for real-time inference control. Experiments show our approach establishes a superior Pareto-frontier for early exit. CUSUM achieves 63.06% accuracy with 11.1% token reduction, outperforming DEER and Dynasor by 3.28% and 4.36% in accuracy respectively. For test-time scaling, CUSUM-weighted voting consistently outperforms self-consistency.

Ting Xu, Xu He, Yupu Lu +4

The Chinese University of Hong Kong · Huawei Technologies Ltd · School of Computing and Data Science, The University of Hong Kong

CoT Reasoning Test-Time Scaling LLM Inference Acceleration Efficient Language Model Inference Reasoning Trace Analysis

Ask PDF

Aug 3, 2026

A Taxonomy of Cognitive Capability Gaps in Generative and Agentic AI

Cognitive AI seeks to move beyond language generation and autonomous task execution toward systems capable of sustained reasoning, adaptive behavior, persistent memory, and self-regulation. While generative and agentic AI have demonstrated impressive capabilities across a wide range of tasks, many fundamental cognitive functions remain fragmented or weakly developed, limiting reliable operation over extended time horizons. This paper presents a taxonomy-driven survey of the major cognitive capability gaps that continue to constrain the development of Cognitive AI. The literature is organized around five dimensions: persistent state modeling, goal-directed autonomy, self-monitoring and control, environment interaction, and learning and adaptation. For each dimension, we review recent advances, identify recurring limitations, and discuss open research challenges. Building on these insights, we outline a conceptual Adaptive Cognitive Intelligence Architecture (ACIA) and examine emerging directions in cognition-centric evaluation. The proposed taxonomy provides a unified framework for organizing existing research, identifying unresolved challenges, and guiding the design of future cognitively capable systems. Together, the taxonomy, architectural perspective, and evaluation framework offer a roadmap for advancing AI systems that exhibit more reliable long-term reasoning, adaptive decision-making, and continual learning. The survey highlights key research opportunities toward more adaptive, reliable, and cognitively capable AI systems, providing a foundation for future progress toward Cognitive AI and, ultimately, Artificial General Intelligence (AGI).

Taye Akinrele, Sindhuja Penchala, Noorbakhsh Amiri Golilarz +2

Department of Computer Science, The University of Alabama, Tuscaloosa, Alabama, USA.

Cognitive Architectures for AI Agents Agentic Reasoning

Ask PDF

Abstract

AI Agent Reliability

Explore similar work

Think Fast: Estimating No-CoT Task-Completion Time Horizons of Frontier AI Models

Unveiling the Entropy Dynamics of Chain-of-Thought Reasoning

A Taxonomy of Cognitive Capability Gaps in Generative and Agentic AI