DLDaniel R. Little

Daniel R. Little

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Daniel R. Little has a total of 143 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Prioritization of Perceptual Decisions

    by Daniel R. Little, Luke Russell, Peizhao Xie, Ruby Steinberg, Asha Bartlett, Ami Eidels

    Scheduling theory provides optimal policies for allocating limited resources to multiple tasks over time. However, little research has examined how cognitive mechanisms perform relative to these optimal policies. We introduce a novel experimental paradigm to study human scheduling decisions using a set of perceptual tasks (random dot kinematograms, RDKs) that vary in difficulty. A foundational question in this paradigm is how optimal prioritization depends on the representation of task factors. Participants selected and completed RDKs one at a time, with the goal of completing as many as possible before a deadline. In Experiment 1 (collected in 2020), RDK difficulty was explicitly labeled. Participants showed near-optimal scheduling, selecting easier tasks first, especially under time pressure. In Experiment 2 (collected in 2022), difficulty had to be inferred from dynamic RDK displays. This resulted in less optimal scheduling, closer to random selection. In Experiment 3 (collected in 2026), we removed observation time from the selection and completion time by limiting the time window in which participants could view the moving RDKs. Across increasing time windows from 200 to 1600 ms, we showed that the optimality of the resulting schedule grew with increasing observation time. We develop a cognitive model of scheduling based on the Plackett-Luce model of ranked preferences. The model captures key differences between experiments in terms of the strength parameters associated with each difficulty level. Our paradigm and modeling approach provide a foundation for studying human scheduling across a range of task types and goals.

  • Open Access

    Right Error Wrong Time: Condition-Dependent Fast and Slow Errors Distort Architecture Inference in Systems Factorial Technology

    by Hang Qian, Paul Michael Garrett, Philip L. Smith, Daniel R. Little

    Systems Factorial Technology infers processing architecture from response-time distributions obtained under factorial manipulations of information quality. These analysis are conventionally restricted to correct responses, leaving unclear whether condition-dependent errors can distort the architecture-diagnostic survivor interaction contrast (SIC). We used diffusion-model simulations to examine whether SIC-based diagnoses remain valid when errors differ in prevalence and timing across the high–high, high–low, low–high, and low–low conditions. We first implemented coactive processing using the partitioned circular diffusion model under AND and OR task rules. Cross-trial variability in drift magnitude produced slow errors, whereas variability in the decision boundary produced fast errors. Although the expected stochastic ordering of the correct-response survivor functions was generally preserved, both mechanisms progressively attenuated the characteristic negative-then-positive coactive SIC, causing it to resemble the entirely positive signature of parallel self-terminating processing. We then generalized the analysis to a parallel self-terminating architecture implemented using linear-drift, linear-infinitesimal-variance model. Racing generated slow errors even without cross-trial drift-rate variability, but the expected positive SIC remained robust while the error pattern was relatively similar across factorial conditions. Adding drift-rate variability differentiated the condition-specific error patterns and progressively weakened the SIC. These findings show that presence of errors compromises architectural diagnosis when their timing and prevalence differ across four factorial conditions, causing correct-response selection to reshape the contributing RT distributions unequally. Conditional accuracy functions provide a practical companion measure for identifying this risk.

  • Open Access

    The Perception and Metacognition of Time Within and Between Stimulus Modalities

    by Aris Yiu, Daniel R. Little, Simon James Cropper

    Time perception depends not only on judging duration but also on evaluating how accurate that judgement was. Our previous work compared visual, auditory, and audiovisual intervals separately and observed a modality asymmetry at short durations. The present study examined whether pairing them within a trial reproduces this pattern, and whether order matters. Eleven participants completed a modified temporal bisection task, bisecting two consecutive intervals then choosing the more accurate. Intervals were both audiovisual (AV), auditory followed by visual (A+V), or visual followed by auditory (V+A), tested at 1500 and 3000 msec. The AV condition ran twice, showing no reliable repetition difference. The modality asymmetry did not reappear once the two modalities were directly paired. Neither Modality nor Duration reliably affected the metacognitively-selected estimate. However, this stability did not extend to metacognitive accuracy, indexed by our Metacognitive Index (MCI), which was higher for V+A at the shorter duration, though only anecdotally (BF10 = 1.43). A hierarchical Bayesian model tracked the observed data closely, with individual consistency explaining much of the between-participant variability. In summary, perceiving a duration and judging one’s own performance may not be the same computation, with the latter more sensitive to modality-induced distortion.

  • Open Access

    Redundancy Repartitions Capacity and Enhances Recall in Visual Working Memory: The Sample-Size Model of Neural Activation

    by Paul Michael Garrett, Yu Liu, Daniel R. Little, Philip L. Smith

    Visual working memory (VWM) has a limited capacity of around four items, yet recall is also improved by structure in the memory array. We reconcile these views by testing how color redundancy – a low-level structural regularity – alters recall precision in a continuous color-reproduction task using a between-subjects (21 participants) and small-N within-subjects design (5 participants). We model precision by combining the attention-weighted sample-size model of Smith et al. (2018) with Oberauer's (2023) signal discrimination model. This sample-size model of neural activation describes baseline predictions for how capacity limits and recall sensitivity should theoretically improve with redundancy, and can be used as a diagnostic process model of VWM and attention. Our model shows that array regularities enhance overall item capacity by storing redundant items as unitary objects and that redundant items often benefit from probability summation. Our model also identifies nuanced individual attentional strategies: trade-offs in redundant and non-redundant item recall, redundant-item neural summation, and Gestalt-like encoding strategies across redundant and non-redundant items. These individual differences would be masked by data aggregation. Our results motivate the use of individual-level modeling when theorizing about nuanced relationships, such as exists between attention and VWM.

  • Open Access

    Attentional Modulation and Processing Architecture in Perceptual Categorization

    by Daniel R. Little, Sarah Moneer

    The processing architecture underlying perceptual categorization can be modulated by the spatial arrangement of features. While separable dimensions are typically processed serially when presented in distinct spatial locations, overlapping or spatially adjacent features may allow for parallel processing. This study investigates how spatial proximity influences the deployment of attention and, consequently, the architecture of cognitive processing. We conducted two experiments that systematically varied the spatial separation of two diagnostic features (saturation and orientation) embedded within schematic stimuli. Using the logical rules paradigm and Systems Factorial Technology (SFT), we assessed whether processing was best described by serial, parallel, or mixed-model architectures. Results revealed that increased spatial separation biased processing toward serial architectures, while overlapping and adjacent features led to mixtures of serial and parallel processing. These findings support the notion that attentional allocation is sensitive to spatial layout, and they highlight how perceptual organization constrains the architecture of information processing during categorization. Implications for models of visual attention and perceptual decision-making are discussed.

  • Open Access

    Descriptor: Appraising Trust in Interactions Between Human Supervisors and Autonomous Robotic Systems (ATHAI)

    by Daniel A. Williams, Airlie Chapman, Daniel R. Little, Chris Manzie

    With the ongoing development of autonomous systems, significant attention has focused on deploying autonomous robotic systems to complete a task under human supervision. An important pillar of this interaction is a notion of human trust in the autonomous system, however the modeling of trust in such scenarios has been hampered by the lack of a standard open-access dataset with longitudinal measurements of trust. By recording measurements of trust at a single point in time (often at the end of interactions), these datasets do not capture the temporal dynamics of trust over the course of interactions. We propose to address these challenges with a novel multi-modal dataset that aggregates autonomous system status and task performance, self-reported human trust, and human interventions in the autonomous system during interactions. We collect the data using a software-based interface that simulates a swarm of autonomous robotic agents performing a foraging task in a rectangular area. By recording metrics pertaining to the autonomous robotic swarm’s task performance and human trust, the dataset provides a foundation for the modeling of human trust dynamics and their influence on human interventions. In addition, the dataset’s inclusion of timestamped human responses to a secondary task permits analysis of human attention allocation between tasks over time. It is anticipated that the dataset will provide a benchmark for human trust models, facilitating the development of more trust-aware autonomous robotic systems. IEEE SOCIETY/COUNCIL Robotics and Automation Society (RAS).

  • Open Access

    Recency is Sufficient for Reconciling Categorisation and Memory: Commentary on Devraj et al. (2024)

    by Daniel Robert Hutchinson, Daniel R. Little, Adam F Osth

    Devraj et al. (Psychonomic Bulletin & Review. doi:10.3758/s13423-023-02448-2, 2024) argued that findings which suggest that memories for items become less accessible over time conflict with categorisation findings where exemplar performance improves across training. To reconcile this, they highlighted that under real-world conditions items tend to reappear less frequently over time, thus preferentially maintaining new items can improve performance. Typical categorisation experiments instead distribute exemplars uniformly across trials. Instead, under a power-law stimulus distribution Devraj et al. showed worsening fit for exemplar classification models across trials. They used this as evidence that forgetting behaviour adapted to task demands, reducing exemplar were accessibility and encouraging prototype use for classification decisions. By re-analysing the same data, we argue instead that this pattern can be produced with exemplar-forgetting in both conditions. By systematically increasing in the delays across which stimuli were tested, their Experimental condition exaggerated the effects of forgetting on performance in later trials compared to the Control condition. This resulted in a reversal of performance growth across trials – instead leading to a steady decline in performance. As exemplar model-fit advantage is expected to vary with performance, we suggest that trends in this advantage are not diagnostic of a shift in classification strategy. We found that a forgetting-function improved exemplar model fit to Devraj et al.’s data, and under reasonable parameters could predict the observed patterns of performance and model-fit a-priori. Compared with a strategy-shifting mixture model, exemplar-forgetting provided equivalent fits despite being more theoretically parsimonious. We suggest power-law memory decay does not produce a tension between categorisation and memory findings, as increased forgetting is found across longer retention intervals whereas the delay between exemplar learning and classification remains constant across typical categorisation experiments.

  • Open Access

    Second-order facial features are processed analytically in composite faces

    by Xue Jun Cheng, Daniel R. Little

    In contrast to claims of holistic processing, upright aligned composite face morphs were recently shown to be processed in the same manner as inverted or misaligned composite face morphs (Cheng et al. 2018. Journal of Experimental Psychology: Learning, Memory and Cognition, 44 , 833–862). In the present paper, we replicate that work, using a set of schematic faces which vary second-order features (e.g., lip height and eye separation) in the top and bottom halves of the schematic face. We find that the present stimuli show the hallmarks of holistic processing in a complete composite face task, but differ from composite face morphs in that the best fitting MDS metric is more commensurate with an assumption of integrality (i.e., Euclidean distance). Nevertheless, we also find that, as with morph faces, the processing of upright aligned and upright misaligned faces is consistent with a mixture of serial and parallel processing. Importantly, we found little evidence of any strong holistic pooling of the top and bottom face halves into a single object. These results remain consistent with the idea that composite faces are not processed differently from other objects with separable dimensions but instead that composite faces allow more parallel processing when aligned than when misaligned. Data and code are available from: http://github.com/knowlabUnimelb/SCHEMATICFACERULES .


  • Open Access

    Modular Serial-Parallel Network for Hierarchical Facial Representations

    by Mario Fific, Daniel R. Little, Cheng-Ta Yang

    Researchers in the field of face perception have long debated the extent to which we process faces holistically (as a whole) versus analytically (focusing on individual features). While some evidence suggests that faces are perceived holistically, research on the neural organization of the visual system and people's subjective experiences indicate that face parts can also be analyzed individually. This view is corroborated by observations of hierarchical object representation in which selective neural populations are fine-tuned to detect specific visual properties ranging from simple features to more complex combination of features. Thus, advances in theories of face perception are met with two major challenges: How can hierarchical face representations be used to integrate holistic and analytic encoding within the same framework? And how can the stages of face processing be integrated with higher-level cognitive processes, such as memory and decision-making, that are recruited during facial perception? We propose a novel computational framework termed the Modular Serial-Parallel Network (MSPN), which synthesizes several perceptual and cognitive approaches including memory representations, signal detection theory, rule-based decision-making, mental architectures (serial and parallel processing), random walks, and process interactivity. MSPN provides a computational modeling account of four stages in face perception: (a) representational (b) decisional, (c) logical-rule implementation, and (d) modular stochastic accrual of information and can account for both choice probabilities and response-time predictions. In a face classification task, MSPN showed an impressive ability in fitting choice response time distributions over other models. MSPN can be used as a tool for further development and refinement of hypotheses in face perception. The analysis of the model’s parameter values, estimated from data, can be used to explore distinct properties of the perceptual and cognitive processes engaged in both analytic and holistic encoding. We conclude by outlining how MSPN could be generalized to other perceptual and cognitive domains.

  • Asymmetrical Trust Modeling for Human-Robot Swarm Interactions

    by Daniel A. Williams, Airlie Chapman, Daniel R. Little, Chris Manzie

    Advances in the control of autonomous systems have accompanied an expansion in the potential applications for autonomous robotic swarms. The success of applications involving humans depends on the quality of interaction with the swarm, particularly the trust that the commander places in the swarm. Absent from the literature is the design of commander trust dynamics that incorporate asymmetric responses to swarm performance. This paper focuses on developing an estimated trust model that employs a switched linear system structure. The identified model is used in a model-based observer that eliminates the need for self-reported trust measurements. Results from a recent user study with 51 participants illustrate considerations during the estimation of population parameters for such nonlinear model-based observers. It is anticipated that such a trust observer can be used to augment communication interfaces for human-swarm interactions in complex environments, leading to better performing systems incorporating humans in the loop.