Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 May 1.
Published in final edited form as: Dev Psychol. 2023 Mar 2;59(5):963–975. doi: 10.1037/dev0001525

Auditory and visual category learning in children and adults

Casey L Roark 1,2, Erica Lescht 1, Amanda Hampton Wray 1,2, Bharath Chandrasekaran 1,2
PMCID: PMC10164074  NIHMSID: NIHMS1873348  PMID: 36862449

Abstract

Categories are fundamental to everyday life and the ability to learn new categories is relevant across the lifespan. Categories are ubiquitous across modalities, supporting complex processes such as object recognition and speech perception. Prior work has proposed that different categories may engage learning systems with unique developmental trajectories. There is a limited understanding of how perceptual and cognitive development influences learning as prior studies have examined separate participants in a single modality. The current study presents a comprehensive assessment of category learning in 8–12-year-old children (12 Female; 34 white, 1 Asian, 1 more than one race; M household income $85-$100K) and 18–61-year-old adults (13 Female; 32 white, 10 Black or African American, 4 Asian, 2 more than one race, 1 other; M household income $40–55K) in a broad sample collected online from the United States. Across multiple sessions, participants learned categories across modalities (auditory, visual) that engage different learning systems (explicit, procedural). Unsurprisingly, adults outperformed children across all tasks. However, this enhanced performance was asymmetrical across categories and modalities. Adults far outperformed children in learning visual explicit categories and auditory procedural categories, with fewer differences across development for other types of categories. Adults’ general benefit over children was due to enhanced information processing, while their superior performance for visual explicit and auditory procedural categories was associated with less cautious correct responses. These results demonstrate an interaction between perceptual and cognitive development that influences learning of categories that may correspond to the development of real-world skills such as speech perception and reading.

Keywords: category learning, perception, cognition, drift-diffusion modeling, decision making, learning

Category learning is a vital skill in human cognition and categories are ubiquitous across modalities. The ability to sort objects into groups supports object recognition in the visual modality (Richler & Palmeri, 2014) and speech perception in the auditory modality (Holt & Lotto, 2010). Across the lifespan, infants learn to distinguish between familiar and unfamiliar faces and voices, children learn species of animals and types of rocks, and adults may need to learn species of birds for a new birdwatching hobby, sounds of a foreign language, or types of wine. Despite this ubiquity, little is understood about how development affects category learning across modalities as prior research has primarily focused on how individuals learn categories in the visual or auditory modalities. In this study, we examine how children and adults learn auditory and visual categories that tap into different learning systems with distinct developmental trajectories.

Auditory and visual category learning

Many theories of perceptual categorization have primarily focused on the visual modality (Ashby et al., 1998; Kruschke, 1992; Love et al., 2004; Nosofsky, 1986). The visual modality is often posed as a representative modality for understanding perceptual categorization but the modality of stimuli being categorized may influence learning mechanisms.

Fundamentally, visual and auditory signals are different – visual objects are stable over time and vary in space and auditory objects vary in time and are stable in space. As such, the demands for information processing are unique across the modalities. Auditory and visual perceptual systems also undergo different patterns in development. Fundamental auditory abilities are thought to be adult-like by the middle of the first year of life, while higher-level processes continue to develop into childhood (Aslin & Smith, 1988; Werner, 2007). For example, native-language speech representations supporting speech perception are not thought to be adultlike until around 12 years old (Hazan & Barrett, 2000; Idemaru & Holt, 2013; Nittrouer, 2004; Nittrouer et al., 1993; Zevin, 2012).

Fundamental visual abilities develop throughout infancy and childhood (Siu & Murphy, 2018). As with audition, higher-level processes involving vision continue to develop into childhood, with some perceptual milestones complete as late as 20 years (Aslin & Smith, 1988; Siu & Murphy, 2018). It is possible that the differences in development of perceptual systems may contribute to developmental changes in perceptual category learning.

Perceptual processing during learning also undergoes developmental changes. While children can extract and encode task-relevant information regardless of whether it is auditory or visual (Robinson & Sloutsky, 2004; Sloutsky & Napolitano, 2003), they may learn information better when presented aurally than visually (Budoff & Quinlan, 1964; McGeoch & Irion, 1952). In contrast, in adults, auditory and visual learning and memory share many similarities and may be governed by similar foundational principles (Nahum et al., 2010; Visscher et al., 2007).

Dual systems models of category learning

In addition to perceptual changes, children also undergo cognitive changes across development. It is possible that the development of cognitive systems may contribute to developmental changes in category learning. Category learning requires learners to generate and test hypotheses about category identity and process feedback to improve future performance. Learning systems supporting hypothesis testing and feedback processing undergo significant changes across development. Specifically, two learning systems have been proposed to support perceptual category learning – explicit and procedural systems (Ashby et al., 1998; Maddox et al., 2013; Yi et al., 2016 see Dunn et al., 2012 for an alternative perspective).

The explicit system optimally supports learning of categories that can be described by rules (e.g., sopranos and tenors differ based on their vocal ranges; alligators and crocodiles differ based on their color and snout shape), while the procedural system optimally supports learning of categories that are difficult to describe verbally (e.g., speech sounds vary across many dimensions that are difficult to describe; cancerous and non-cancerous moles look different, but these differences cannot be described with simple rules). It is important to note that learning via the procedural system may still occur even when the learning task itself is overt (i.e., supervised). That is, learning via the procedural system does not necessarily need to be implicit (Ashby & Casale, 2002).

Critically, explicit and procedural learning systems have distinct developmental trajectories leading to distinct predictions for differences in learning across development. Explicit learning relies on working memory and selective attention, supported by the prefrontal cortex (PFC), anterior cingulate cortex, and the head of the caudate nucleus in the striatum (Ashby et al., 1998; Ashby & Ell, 2001). The PFC undergoes protracted development and continues developing even into early adulthood (Diamond, 2002; Gogtay et al., 2004; Kolk & Rakic, 2021). The PFC is thought to underlie individual differences and developmental changes in working memory (Cowan, 2016; Gathercole, 1999), selective attention (Plude et al., 1994), impulsivity (Kim & Lee, 2011), perseveration (Hauser, 1999; Munakata et al., 2003), and feedback processing (Di Bernardi Luft, 2014; Ferdinand et al., 2016).

Procedural learning involves learning stimulus-response associations through connections with sensory regions and the body and tail of the caudate nucleus and the putamen (Ashby et al., 1998; Ashby & Ennis, 2006; Ashby & Waldron, 2000). Unlike the PFC, the caudate is thought to be fully adultlike relatively early, around 7 years of age (Casey et al., 2004), with general procedural learning mechanisms becoming adultlike around 10 years of age (Diamond, 2002).

In line with protracted PFC development, learning of rule-based (RB) categories, optimally supported by the explicit system, is better in adults than children (Huang-Pollock et al., 2011; Rabi & Minda, 2014b) and generally improves with age. Adults learn RB categories better than adolescents (13–19-years-old), who themselves learn better than children (7–12-years-old; (Reetzke et al., 2016). To learn RB categories, learners must selectively attend to dimensions that are relevant for categorization and ignore dimensions that are irrelevant. Children’s poorer ability to attend to relevant information and ignore irrelevant information (Deng & Sloutsky, 2015; Geffen & Sexton, 1978; Lane, 1982; Pick et al., 1972; Plebanek & Sloutsky, 2017; Sloutsky, 2010; Strutt et al., 1975) is thought to underlie their poorer RB learning (Rabi et al., 2015; Rabi & Minda, 2014a).

In contrast, the development of learning of information-integration (II) categories, optimally supported by the procedural system, is less clear. Some studies demonstrate, in line with early development of procedural learning abilities, that 3–8-year-old children can be just as accurate as adults in learning visual II categories (Minda et al., 2008; Rabi et al., 2015; Rabi & Minda, 2014b). However, other studies demonstrate that adults learn II categories better than 5–7-year-old (Roark & Holt, 2019) or 8–12-year-old children (Huang-Pollock et al., 2011).

In all, prior work has demonstrated differences in separate groups of children and adults in category learning in a single modality (auditory or visual) and a single category type (RB or II). As a result, it is still unclear how development influences learning of RB and II categories across modalities. In the current study, we comprehensively examine learning across modalities and category types in the same learners.

Decision processes mediating learning

Prior studies have focused on differences in the ability of children and adults to learn perceptual categories. However, little is understood about what underlies these differences. Using drift diffusion models (DDMs) (Ratcliff, 1978; Ratcliff et al., 2016), we will investigate perceptual and cognitive components of decision-making during learning and how this affects performance across development. DDMs assume that evidence is accumulated towards decision alternatives at varying rates and a decision is made once the evidence being accumulated passes a threshold for a specific choice. In the context of category learning, participants see or hear a stimulus that must be categorized, accumulate evidence towards the possible category choices, and make a decision when the evidence for a particular category crosses a participant’s internal decision threshold.

Evidence accumulation rates

Evidence accumulation rates reflect the quality of evidence extracted from a stimulus. With faster accumulation rates, participants process information efficiently, quickly extracting relevant information. As a result, difficulty in perceptual processing is associated with lower evidence accumulation rates (Voss et al., 2004). The evidence accumulation process involves a frontoparietal network (Mulder et al., 2014), including dorsolateral PFC (Rolls et al., 2010). Higher working memory ability is associated with faster evidence accumulation rates (Ester et al., 2014).

Decision thresholds

Decision thresholds reflect the amount of evidence that is accumulated before a decision is made and reflect the speed-accuracy tradeoff (Bogacz et al., 2010). Decision thresholds involve a fronto-basal ganglia network (Mulder et al., 2014) which is relevant for category learning (Ashby et al., 1998; Feng et al., 2021). Higher decision thresholds reflect more cautious responses as individuals wait to gather sufficient information before making a decision.

Decision making processes reflected in these parameters change with development. The few studies that have investigated these processes demonstrate that children from 3–12 years both extract lower quality evidence (i.e., lower evidence accumulation rates) and adopt more conservative decision criteria (i.e., higher thresholds) than adults (Manning et al., 2021; Nordmeyer et al., 2016; Ratcliff et al., 2012; Schneider & Frank, 2016). Here, we will examine decision processes in category learning tasks across modalities to understand how differences in decision processes may depend on the type of information being processed.

Summary

We investigate learning of RB and II auditory and visual categories in 8–12-year-old children and adults to assess whether the nature of the information being processed affects learning and decision-making processes differently across development. To anticipate, we find that adults have an asymmetric benefit over children for auditory-II and visual-RB categories, supported by enhanced information processing efficiency and less cautious correct responses.

Methods

Participants completed three sessions – a general assessment session and two category learning sessions (Figure 1B). Participants completed sessions virtually through the Gorilla Experiment Builder (Anwyl-Irvine et al., 2019) and received $10/hour for participation. Families of the children received an additional $10 for completing all assessments. Participants were all located in the United States.

Figure 1.

Figure 1

Open in a new tab

Stimulus distributions and task procedure

Note. A. Auditory and visual category distributions during training (colored dots) and test (black Xs). B. Study procedure across sessions and within the category learning sessions (2 and 3).

Participants

Children.

Participants were 36 children (12 Female, 24 Male; 34 white, 1 Asian, 1 more than one race) ages 8–12 years (M = 9.97, SD = 1.27) recruited from participation in previous studies or through a local recruitment database. We focus on the 8–12-year age range because this is a period of development of perceptual and cognitive abilities. Nineteen additional children started at least one session but were excluded because they did not complete all tasks (N = 10), were under 8 years old (N = 5) or took more than 2 hours to complete any individual task (N = 4, Mdn = 20.2 min).

Adults.

Participants were 49 adults (13 Female, 35 Male, 1 Non-Binary; 32 white, 10 Black or African American, 4 Asian, 2 more than one race, 1 other or unknown), ages 18–61 years (M = 33.7, SD = 10.6) recruited from Prolific (www.prolific.co). The adult data are part of a larger database for comparison in another study with clinical populations. Twenty-eight participants were selected for this other study based on demographics matching with the clinical population and were excluded from the current study. An additional 45 adults started at least one session but were excluded because they did not complete all tasks. No adults took more than 2 hours to complete any individual task (Mdn = 18.3 min).

A power analysis using the WebPower package in R (Zhang & Yuan, 2018) indicated that in the comparison of performance across age groups and tasks using a mixed measures ANOVA, a sample size of 18 would be needed to detect an effect size of f = 1.04 (adults vs. children effect size from Roark & Holt, 2019) with 95% power and alpha = 0.05. We expected some attrition and so recruited more the than the target sample size. We also expected greater attrition in our adult sample and so we recruited more adults in the first session, leading to more adults in the final sample.

Stimuli

Auditory category stimuli were nonspeech ripple sounds that varied in temporal modulation and spectral modulation. Visual category stimuli were Gabor patches that varied in spatial frequency and orientation. We selected pairs of dimensions across modalities that are the building blocks of auditory and visual perception and have been proposed to be analogs of one another (Visscher et al., 2007).

To allow for comparison across modalities, we created comparable category distributions (Figure 1A). The stimulus distributions were first created in a normalized space and then separately transformed to auditory and visual spaces using the same procedures as a prior study (Roark, Paulon, et al., 2021). The rule-based (RB) categories can be separated based on a unidimensional rule along the temporal modulation (auditory) and spatial frequency (visual) dimensions. The information-integration (II) categories require both dimensions to separate the categories – using only a rule along one dimension would lead to suboptimal performance. Each category type had 200 stimuli (100/category). An additional grid with 64 stimuli was presented in the generalization test.

Procedure

In session 1, adults individually and children’s guardians completed demographic questionnaires, questionnaires on communication and psychological disorders (see Supplementary Materials), and an assessment of working memory capacity. In the second and third sessions participants completed four category learning tasks – RB auditory, RB visual, II auditory, and II visual – and sessions were separated by at least one week to minimize potential carryover effects (Children: M = 15.7 days, SD = 15.3, Mdn = 9.94; Adults: M = 12.2 days, SD = 1.25, Mdn = 12.0). Modality and category were counterbalanced across participants and participants always completed one auditory and one visual task in each session.

A cover story for each task ensure that the study was child-friendly. Participants were told they would hear sounds from two aliens (auditory task) and see crystal balls from two wizards (visual task). Within each task, participants completed four 50-trial blocks of training followed by one 64-trial generalization block where they encountered novel category exemplars and no longer received feedback. On each trial, participants heard a sound or saw an image (1 sec), made an untimed response about the category identity (1 or 2 on the keyboard), and received feedback immediately after their response (smiling face icon for correct and neutral face icon for incorrect), followed by a 1 sec inter-trial interval. Participants were told to be as accurate as possible.

As an index of working memory capacity, we used a task adapted from Luck and Vogel (1997) and Vogel et al. (2001) and based on a child-friendly paradigm by Isbell et al. (2015). Participants were told to detect changes between the appearance of ladybugs on the screen. Ladybugs could appear in one of nine colors (red, pink, purple, blue, green, yellow, orange, brown, and black) on the white background and 2, 4, or 6 ladybugs appeared on the screen at any time. A fixation cross was always present in the center of the screen. Images were created against a 4×3 grid with images pseudo-randomly distributed on the grid such that no consecutive images were in the same exact location or included the exact same ladybug colors.

There were three blocks that varied in the length of presentation of the target and probe arrays – slow (1000 ms), medium (500 ms), and fast (100 ms). Blocks were always completed in this fixed order (slow, medium, fast). Trials of different set sizes were presented randomly within each block (28 trials set size 2, 28 trials set size 4, 28 trials set size 6). After the target image was presented, there was a fixed retention interval of 1000 ms, followed by the probe array. After the final array disappeared, participants were probed to respond whether the images were the same colors (same) or whether one was a different color (different). On different trials exactly one color changed. Half of the trials were same trials and half were different trials. There was an inter-trial interval of 900 ms. Prior to the first block, children completed five practice trials, each followed by immediate feedback (Correct, Incorrect). Feedback was not presented in the rest of the task.

To compute visual working memory capacity (K), we took the same approach as prior research (Isbell et al., 2015; Cowan, 2001): K = SetSize * (Hits – FalseAlarms). We report working memory capacity as the averaged K across set sizes of 4 and 6, as set size 2 may underestimate working memory capacity (Rouder et al., 2008). The working memory task was administered through PsyToolkit (Stoet, 2010; 2017). As expected, adults (M = 2.61) had significantly higher working memory capacity than children (M = 1.82; t(51.0) = 2.74, p = .0084, d = 0.63, 95% CI [0.21, 1.36]).

Drift diffusion modeling

To understand the decision processes involved in learning in children and adults across modalities and types of tasks, we applied drift diffusion models (Ratcliff, 1978; Ratcliff et al., 2016) using a recently developed method of longitudinal drift diffusion mixed modeling using a Bayesian framework (Paulon et al., 2020). We estimated the posterior probabilities of the evidence accumulation rate (i.e., drift) parameter and decision threshold (i.e., boundary) parameter for each category and response combination (e.g., category 1 and response 2) for each participant, task, and block.

We ran the models in the lddmm package in R (Paulon & Sarkar, 2021), first removing the longest 1% and shortest 1% of trials based on RT, and then running the models separately for each task and age group. We ran 6000 iterations of the Markov chain Monte Carlo simulations with a burn in rate of 2000 and a thinning factor of 5.

Transparency and Openness

We reported how we determined our sample sizes, all data exclusions, all manipulations, and all measures in the study, and we follow JARS (Kazak, 2018). Stimulus materials, data, and code for the drift diffusion models are publicly available through the Open Science Framework and can be accessed at https://osf.io/bmk9v/?view_only=24afcac4a8704bca968ca24890c60c1f. The data were analyzed using R, version 3.6.1 (R Core Team, 2019) and visualizations were created using the R packages ggplot2, version 3.3.5 (Wickham, 2016) and ggthemes, version 4.2.0 (Arnold, 2018). The study was not preregistered.

Results

To understand auditory and visual RB and II category learning in children and adults, we compared category learning performance (accuracies, reaction times) and decision processes (i.e., evidence accumulation rates and decision thresholds) during learning across the four tasks.

Category learning

Both children and adults successfully learned the categories (Figure 2A), with significantly above-chance (50%) performance in all tasks (one-sample t-tests, all ps < .001) and, on average, children were slower than adults (Figure 2B; t(213.8) = 6.08, p < .001, d = 0.69; 95% CI [208, 407], MChildren = 910 ms, MAdults = 603 ms). Adults and children also successfully transferred their performance to stimuli not encountered during training (Supplementary Materials). We next examine differences between adults and children in category learning accuracies.

Figure 2.

Figure 2

Open in a new tab

Category learning accuracies and reaction times across blocks and in generalization test

Note. A. Category learning accuracies across tasks in children and adults. Percentages shown are the average differences between age groups in block 1 and block 4. B. Reaction times across tasks in children and adults. Dashed lines reflect visual modality. Error bars reflect SEM.

We compared accuracies across blocks, modalities, categories, and age groups using an omnibus mixed model ANCOVA with working memory score (K) as a covariate. After adjusting for working memory, we found a significant three-way interaction between category, modality, and age group (F(1, 82) = 10.6, p = .0020, ηG2 = 0.015). Below, we report the results from follow up analyses to unpack this interaction using additional ANCOVAs.

Children’s performance across tasks.

Among children, after accounting for working memory scores, there were no significant differences in performance across categories or modalities (F(1, 34) = 3.13, p = .086, ηG2 = 0.012; MVisual-RB = 61%; MVisual-II = 62%; MAuditory-RB = 61%; MAuditory-II = 58%). To understand whether performance changed with age among the children, we examined the relationship between age (in decimal years) and final block accuracy. There was no significant relationship between age and final block accuracy in any of the tasks (βAge*II-Auditory = −0.013, SE = 0.018, p = .46; βAge*II-Visual = −0.0067, SE = 0.021, p = .75; βAge*RB-Auditory = −0.032, SE = 0.021, p = .13; βAge*RB-Visual = −0.013, SE = 0.021, p = .53).

Adults’ performance across tasks.

In contrast to children, adults learned some categories better than others, with substantially better accuracy for the visual-RB categories (M = 75%) than other types of categories (MVisual-II = 68%; MAuditory-RB = 68%; MAuditory-II = 70%). Among adults, there was a significant interaction between category and modality after accounting for working memory (F(1, 47) = 8.75, p = .0050, ηG2 = 0.029). Adults learned the Visual-RB categories significantly better than the Visual-II categories (p < .001) but there were no differences in performance for the Auditory-RB and Auditory-II categories (p = .38). There was no significant relationship between age and final block accuracy in any of the tasks for adults (βAge*II-Auditory = 0.0021, SE = 0.0019, p = .27; βAge*II-Visual = −0.00041, SE = 0.0020, p = .84; βAge*RB-Auditory = −0.0020, SE =0.0020, p = .31; βAge*RB-Visual = −0.00029, SE = 0.0020, p = .88).

Comparing children’s and adults’ performance.

Even when accounting for differences in working memory, children had poorer accuracies than adults across all four tasks. However, the improvement with age was asymmetrical across categories and modalities. We compared performance across children and adults by running separate mixed model ANCOVAs for each modality with category and age group as factors and working memory score as a covariate.

For both the auditory and visual modalities, there was a significant interaction between category and age group (Auditory: F(1, 82) = 4.09, p = .046, ηG2 = 0.014; Visual: F(1, 82) = 8.69, p = .0040, ηG2 = .025). This interaction was due to differences in the size of the advantage adults had over children for the RB and II categories. Adults far outperformed children for the auditory-II categories (M = 12%, 95% CI [7.79, 17.1], p < .001) with a smaller advantage for the auditory-RB categories (M = 7%, 95% CI [1.76, 12.9], p = .028). The pattern in the visual modality was reversed. Adults far outperformed children for the visual-RB categories (M = 14%, 95% CI [8.43, 19.6], p < .001) with a smaller advantage for the visual-II categories (M = 7%, 95% CI [1.89, 11.1], p = .024). In summary, while adults outperformed children across all tasks, the size of this difference in performance depended on the interaction of modality and category. Specifically, adults far outperformed children in auditory-II and visual-RB tasks and less strongly outperformed children in the auditory-RB and visual-II tasks.

Decision processes

Adults were both significantly faster and more accurate than children in all four tasks, with especially superior learning in auditory-II and visual-RB tasks. What enables this asymmetric enhanced performance? To answer this question, we applied DDMs, which use both accuracy and reaction time to estimate the perceptual and cognitive processes underlying decision making. As a reminder, higher evidence accumulation rates reflect increased efficiency of information processing and higher decision thresholds indicate more cautious responses, favoring accuracy over speed. Importantly, differences in reaction times may be due to differences in evidence accumulation rates, decision thresholds, or both. For example, with the same evidence accumulation rate, longer RTs will reflect higher thresholds, while given the same decision threshold, longer RTs will reflect lower evidence accumulation rates. When both measures vary, longer RTs can be primarily driven by one or the other variable or both. These models use Bayesian analyses and so we interpret differences between groups where there is no overlap between the 95% credible intervals (Figures 3–4), which reflect the 95% posterior probability estimates for each parameter.

Figure 3.

Figure 3

Open in a new tab

Evidence accumulation rates across blocks

Note. A. Evidence accumulation rates for correct responses in children and adults shown separately for each task. B. Evidence accumulation rates for correct responses minus rates for incorrect responses. Values near zero (dotted line) reflect no differences between correct and incorrect responses, negative values reflect faster accumulation rates for incorrect responses than correct responses, and positive values reflect faster accumulation rates for correct than incorrect responses. Dashed lines reflect visual modality. Error bars reflect 95% credible intervals.

Figure 4.

Figure 4

Open in a new tab

Decision thresholds across blocks

Note. A. Decision thresholds for correct responses for children and adults shown separately for each task. B. Decision thresholds for correct responses minus thresholds for incorrect responses. Values near zero (dotted line) reflect no differences between correct and incorrect responses, negative values reflect less cautious thresholds for correct responses than incorrect responses, and positive values reflect more cautious thresholds for correct than incorrect responses. Dashed lines reflect visual modality. Error bars reflect 95% credible intervals.

Efficiency of evidence accumulation.

Children had small and flat evidence accumulation rates across all four tasks (Figure 3A). Initially, children had higher rates for auditory-II task than the other tasks, but by the end of training, all four tasks had similar evidence accumulation rates. Adults initially had higher rates for auditory than visual tasks, but this disappeared over the course of learning. Overall, in both children and adults, evidence accumulation rates were similar across different tasks. This reflects general stability of perceptual processes and the ability to extract information from stimuli in these tasks.

Across all four tasks and in all blocks, adults had higher evidence accumulation rates than children. This efficient information processing may support adults’ generally superior performance across all tasks. Children may have poorer performance because they are inefficient at extracting relevant information to guide their decisions. To better contextualize these patterns, we examined the difference in evidence accumulation rates for correct and incorrect responses (Figure 3B). This comparison demonstrates that adults have more efficient information processing for correct than incorrect responses in all tasks, whereas children had similar patterns of evidence accumulation for correct and incorrect responses. Adults were more successful than children across all tasks because they were able to quickly and efficiently accumulate evidence toward the correct category choice throughout training. This pattern does not explain why adults were so much better than children at auditory-II and visual-RB categories, as evidence accumulation rates were similar across tasks.

To understand whether evidence accumulation rate varied with age within the two age groups, we examined the relationship between age and final block accumulation rate. There was no significant relationship between age and evidence accumulation rate in any of the tasks for children (βAge*II-Auditory = 0.00046, SE = 0.0047, p = .92; βAge*II-Visual = −0.00046, SE = 0.0066, p = .94; βAge*RB-Auditory = 0.000053, SE = 0.0066, p = .99; βAge*RB-Visual = 0.0045, SE = 0.0066, p = .50) or adults (βAge*II-Auditory = 0.010, SE = 0.010, p = .31; βAge*II-Visual = −0.013, SE = 0.014, p = .37; βAge*RB-Auditory = −0.0075, SE = 0. 014, p = .60; βAge*RB-Visual = 0.0016, SE = 0. 014, p = .91).

Decision thresholds.

Unlike evidence accumulation rates, children’s decision thresholds differed across tasks (Figure 4A). Children had higher thresholds in the auditory than the visual tasks and in the II than the RB tasks. In other words, children waited to gather more information about the auditory and II stimuli before making a decision. Adults initially also had more cautious responses for auditory than visual tasks and had especially lower thresholds for the task they learned the best (visual-RB).

Across all blocks of all tasks except the final block for the auditory-II task, adults had higher thresholds than children. To contextualize these patterns, we examined the difference in decision thresholds for correct and incorrect responses (Figure 4B). Larger differences between correct and incorrect response thresholds reflect differentiation between these response types, with negative values indicating that participants gathered less information and responded faster when they made a correct decision relative to an incorrect decision. Children and adults both had lower thresholds for correct than incorrect responses, but the magnitude of this difference varied across the four tasks. For tasks in which the adults’ performance greatly exceeded children’s performance (auditory-II, visual-RB), adults had much larger differences in the thresholds for correct and incorrect responses than children. In contrast, for the other two tasks (auditory-RB, visual-II), children and adults had similar patterns of decision thresholds. In other words, for the auditory-II and visual-RB tasks, adults made faster, more accurate responses because they gathered evidence much more efficiently than children, while also needing less information before making a correct decision.

Just as with evidence accumulation rate, there was no significant relationship between age and decision thresholds in any of the tasks for children (βAge*II-Auditory = −0.024, SE = 0.027, p = .39; βAge*II-Visual = 0.062, SE = 0.036, p = .09; βAge*RB-Auditory = 0.0098, SE = 0.036, p = .79; βAge*RB-Visual = 0.046, SE = 0.036, p = .29) or adults (βAge*II-Auditory = −0.00068, SE = 0.0024, p = .78; βAge*II-Visual = −0.00018, SE = 0.0033, p = .96; βAge*RB-Auditory = 0.00038, SE = 0.0033, p = .91; βAge*RB-Visual = 0.0015, SE = 0.0033, p = .65).

Discussion

We examined auditory and visual learning in 8–12-year-old children and adults across dissociable category structures. Our goal was to evaluate changes in category learning as a function of age. Critically, all participants learned all categories, allowing for systematic examination of the impacts of modality (auditory, visual) and category (rule-based, information-integration) on learning and decision processes in the same individual. Overall, we found that adults learned better and responded faster than children for all categories across modalities. However, there were significant asymmetries across development. Adults showed especially enhanced learning relative to children for visual categories that that could be described by a simple, unidimensional rule and for auditory categories that could not be described by a simple rule. These results provide substantial insights into the nature of perceptual category learning in children and adults and highlight the need to understand the role of modality in learning.

Adults learned better than children even when controlling for working memory

In line with prior work separately examining visual (Huang-Pollock et al., 2011; Rabi & Minda, 2014a; Minda et al., 2008) and auditory category learning (Reetzke et al., 2016; Roark & Holt, 2019), we found that adults outperformed children for all types of categories. We also found that children were generally slower than adults, replicating a general finding across research domains (e.g., Manning et al., 2021; Ratcliff et al., 2012).

Moving beyond prior studies, using drift diffusion models, we found that the nature of differences in learning performance in children and adults was due to differences in decision-making processes across development. Across modalities and categories, we found that adults have more efficient information processing than children. Adults’ higher evidence accumulation compared to children is similar to prior work in other tasks (Manning et al., 2021; Nordmeyer et al., 2016; Ratcliff et al., 2012; Schneider & Frank, 2016). This finding also complements prior work in adults, where evidence accumulation rates are associated with individual differences in working memory abilities (Ester et al., 2014). Importantly, differences in performance between adults and children were present even after accounting for differences in working memory ability. This indicates that developmental differences may stem from multiple sources including but not limited to the development of cognitive abilities like working memory. For instance, higher evidence accumulation rates are also related to enhanced attention and motivation (Gesiarz et al., 2019; Ritz et al., 2020; Voss et al., 2004), which is likely higher in adults.

Supporting the view that adults have enhanced task-based attention and motivation, we found that while adults’ performance continued increasing across blocks, children stalled in learning and did not improve after the first block. The differences in these patterns may reflect differences in the goals of children and adults. Adults seem more motivated than children to maximize performance in these tasks. This interpretation is in line with results from reinforcement learning tasks where adults consistently use strategies that maximize reward, but children do not (Blanco & Sloutsky, 2019, 2020, 2021; Gopnik, 2020; Liquin & Gopnik, 2022).

We also did not find significant age-related changes when examining children and adults separately. As category learning itself has wide variability regardless of age, it may be difficult to observe a possible relationship between age and performance with a relatively small sample (Children N = 37, Adults N = 51). Future work should examine the relationship between learning and age across a wide range (including adolescence) with a very large sample to be able to conclude when age-related changes in category learning occur. With this caveat, we observed large and broad differences in learning processes and outcomes in 8–12-year-old children and 18–61-year- old adults, that were relatively stable within these age ranges.

Asymmetric development across categories and modalities

Even though adults outperformed children across tasks, there were asymmetries in the differences between children and adults across different types of categories. Adults’ performance benefits over children were especially large for auditory-II and visual-RB categories. Adults excelled in the auditory-II and visual-RB tasks specifically because adults efficiently gathered information while maintaining cautious decision thresholds. Even though adults generally had shorter reaction times compared to children, their performance is supported by very efficient evidence accumulation rates despite their overall higher decision thresholds.

Prior work has generally demonstrated that children have more cautious responses than adults (Manning et al., 2021; Nordmeyer et al., 2016; Ratcliff et al., 2012; Schneider & Frank, 2016), though in some tasks children make more impulsive, less cautious decisions than adults (Martinez et al., 2018). Here, we found that, in general, children had less cautious responses than adults and that children had more similar decision thresholds for correct and incorrect responses. This indicates that children were more likely to reach an incorrect threshold than adults because they did not learn to adjust the speed of their responses for correct or incorrect choices. In contrast, adults gathered less information for correct choices and more information for incorrect choices. This pattern was limited to the categorization problems that adults particularly excelled in relative to children – auditory-II and visual-RB.

We posit that auditory-II and visual-RB learning are exceptional in adults relative to children because of developmental changes in modality-specific processing demands. Auditory information processing may privilege more non-analytic processing (aligned with II tasks), whereas visual information processing may privilege more analytic processing (aligned with RB tasks). While auditory signals are relatively more difficult to break down into their component parts and can be described as more integral than separable, many visual signals are easily separable (Garner, 1974). As a result, it may be more difficult to selectively attend to individual components in an auditory signal than a visual signal (such as during RB categorization). This pattern could affect performance across development as there is a tendency for young children to perceive stimuli across modalities as integral while adults can easily selectively attend to separable dimensions (Kemler & Smith, 1978; Smith & Kemler, 1978). This shift from integrality to separability is thought to occur between 5 and 8 years of age (Kemler & Smith, 1978) so this developmental difference may not account for differences among adults and children in the current study. It is also important to consider that the world is multimodal, with single category identity often cued by information from auditory and visual modalities simultaneously and work has shown that children learn from unimodal and multimodal signals differently (Broadbent et al., 2018; Kirkham et al., 2019).

Prior work has also demonstrated that children have enhanced learning for auditory over visual information (Budoff & Quinlan, 1964), whereas adults have enhanced learning for visual over auditory information (Roark, Smayda, et al., 2021; Sloutsky & Napolitano, 2003). We found that children do not have modality bias for category learning and performed similarly across modalities. Adults also did not have a modality bias for category learning and instead showed enhanced visual-RB learning relative to the three other types of categories. These results demonstrate that developmental differences for category learning tasks cannot be defined based on modality or category type alone. Instead, these factors must be considered together. Although adults were better at learning than children in all four tasks, adults were especially superior learners than children in auditory-II and visual-RB tasks.

The specific advantages of adults over children in auditory-II and visual-RB category learning may also relate to developmentally relevant experiences in the auditory and visual modalities. With age, expertise develops in particular domains of auditory and visual modalities – speech and reading. Speech perception and reading are both inherently multimodal processes. However, speech perception can occur robustly in the absence of visual information and reading can occur in adults without the need for explicit direct mapping to auditory information. That is, at expert levels, speech can be understood as a primarily auditory signal and reading of text can be understood as a primarily visual signal.

Children typically become proficient readers around ages 7–8 years and continue developing reading abilities throughout childhood and adolescence (Chall, 1983; Horowitz-Kraus et al., 2017; Kuhn & Stahl, 2003; National Reading Panel, 2000; Nippold, 2006; Rasinski & Hoffman, 2003). While children have proficient representations of native language speech sounds in the first year of life (Werker & Tees, 1984), they continue developing these representations at least until age 12 (Hazan & Barrett, 2000; Idemaru & Holt, 2013; Nittrouer, 2004; Nittrouer et al., 1993; Zevin, 2012). As these are both very complex and important species-specific skills, it is likely that development of speech perception and reading impact children’s development and performance in other domains. It is possible that adults have outsized benefits over children for auditory-II and visual-RB learning because these learning tasks reflect the same kind of processing required for complex problems at which adults have become experts as a function of their lifelong experience. That is, listening to natural speech may reflect an auditory-II type problem, with multiple dimensions that are difficult to verbalize, and reading may reflect a visual-RB type problem, with fairly consistent letter-to-phoneme mapping.

Conclusion

Prior investigations of perceptual category learning have primarily focused on a single modality at a time, with most work examining visual categories. Even so, category learning is a critical skill across modalities and the lifespan. The current study demonstrates the importance of consideration of both perceptual and cognitive development and the ways in which they influence how learners approach different learning tasks across modalities. While adults generally outperform children in auditory and visual learning, the differences between children and adults were larger for auditory-II and visual-RB categories. These patterns may align with the specific types of expertise that adults have in native language speech perception and reading that are still developing in children. These results have implications for understanding how perceptual and cognitive changes across development influence learning.

Supplementary Material

Supplemental Material

Public significance statement:

Category learning is a fundamental ability that supports complex processes across modalities and across the lifespan. We investigate learning of auditory and visual categories in 8–12-year-old children and adults. Categories that relate to real-world learning challenges children encounter in speech perception and reading are learned much better in adults than children and are associated with distinct decision processes.

Author note:

This research was supported by the National Institute on Deafness and Other Communication Disorders (F32DC018979 to C.L.R. and R21DC017227 to A.H.W.). Stimulus materials, data, and code for the drift diffusion models are publicly available through the Open Science Framework and can be accessed at https://doi.org/10.17605/OSF.IO/BMK9V. The study was not preregistered.

References

  1. Anwyl-Irvine A, Massonnié J, Flitton A, Kirkham N, & Evershed J (2019). Gorilla in our Midst: An online behavioral experiment builder. Behavior Research Methods, 438242. 10.3758/s13428-019-01237-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arnold JB (2018). ggthemes: Extra Themes, Scales and Geoms for “ggplot2” (4.2.0) [R package]. https://CRAN.R-project.org/package=ggthemes
  3. Ashby FG, Alfonso-Reese LA, Turken AU, & Waldron EM (1998). A neuropsychological theory of multiple systems in category learning. Psychological Review, 105(3), 442–481. 10.1037/0033-295x.105.3.442 [DOI] [PubMed] [Google Scholar]
  4. Ashby FG, & Casale MB (2002). The cognitive neuroscience of implicit category learning. Attention and Implicit Learning, 109–141. 10.1016/j.neubiorev.2007.11.002 [DOI] [Google Scholar]
  5. Ashby FG, & Ell SW (2001). The neurobiology of human category learning. Trends in Cognitive Sciences, 5(5), 204–210. 10.1016/s1364-6613(00)01624-7 [DOI] [PubMed] [Google Scholar]
  6. Ashby FG, & Ennis JM (2006). The role of the basal ganglia in category learning. In Ross BH (Ed.), The Psychology of Learning and Motivation (Vol. 46, pp. 1–36). Elsevier. [Google Scholar]
  7. Ashby FG, & Waldron EM (2000). The Neuropsychological Bases of Category Learning. Current Directions in Psychological Science, 9(1), 10–14. 10.1111/1467-8721.00049 [DOI] [Google Scholar]
  8. Aslin RN, & Smith LB (1988). Perceptual development. Annual Review of Psychology, 39, 435–473. [DOI] [PubMed] [Google Scholar]
  9. Blanco NJ, & Sloutsky VM (2019). Adaptive Flexibility in Category Learning? Young Children Exhibit Smaller Costs of Selective Attention Than Adults. Developmental Psychology, October. 10.1037/dev0000777 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Blanco NJ, & Sloutsky VM (2020). Attentional mechanisms drive systematic exploration in young children. Cognition, 202, 104327. 10.1016/j.cognition.2020.104327 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Blanco NJ, & Sloutsky VM (2021). Attentional mechanisms drive systematic exploration in young children. Cognition. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Bogacz R, Wagenmakers E-J, Forstmann BU, & Nieuwenhuis S (2010). The neural basis of the speed–accuracy tradeoff. Trends in Neurosciences, 33(1), 10–16. 10.1016/j.tins.2009.09.002 [DOI] [PubMed] [Google Scholar]
  13. Broadbent HJ, Osborne T, Rea M, Peng A, Mareschal D, & Kirkham NZ (2018). Incidental Category Learning and Cognitive Load in a Multisensory Environment Across Childhood. Developmental Psychology, 54(6), 1020–1028. 10.1037/dev0000472 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Budoff M, & Quinlan D (1964). Auditory and Visual Learning in Primary Grade Children. Child Development, 35(3), 583–586. 10.1111/j.1467-8624.1964.tb05196.x [DOI] [PubMed] [Google Scholar]
  15. Casey BJ, Davidson MC, Hara Y, Thomas KM, Martinez A, Galvan A, Halperin JM, Rodríguez-Aranda CE, & Tottenham N (2004). Early development of subcortical regions involved in non-cued attention switching. Developmental Science, 7(5), 534–542. 10.1111/j.1467-7687.2004.00377.x [DOI] [PubMed] [Google Scholar]
  16. Chall JS (1983). Stages of reading development. McGraw-Hill. [Google Scholar]
  17. Cowan N (2001). The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioral and Brain Sciences, 24(1), 87–114. 10.1017/s0140525×01003922 [DOI] [PubMed] [Google Scholar]
  18. Cowan N (2016). Working Memory Maturation. Perspectives on Psychological Science, 11(2), 239–264. 10.1177/1745691615621279 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Deng WS, & Sloutsky VM (2015). The Development of Categorization: Effects of Classification and Inference Training on Category Representation. Developmental Psychology, 1–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Diamond A (2002). Normal Development of Prefrontal Cortex from Birth to Young Adulthood: Cognitive Functions, Anatomy, and Biochemistry ([“Stags” & “Knight”], Eds; pp. 466–503).
  21. Di-Bernardi-Luft C (2014). Learning from feedback: The neural mechanisms of feedback processing facilitating better performance. Behavioural Brain Research, 261, 356–368. 10.1016/j.bbr.2013.12.043 [DOI] [PubMed] [Google Scholar]
  22. Dunn JC, Newell BR, & Kalish ML (2012). The effect of feedback delay and feedback type on perceptual category learning: The limits of multiple systems. Journal of Experimental Psychology: Learning, Memory, and Cognition, 38(4), 840–859. 10.1037/a0027867 [DOI] [PubMed] [Google Scholar]
  23. Ester EF, Ho TC, Brown SD, & Serences JT (2014). Variability in visual working memory ability limits the efficiency of perceptual decision making. Journal of Vision, 14(4), 2–2. 10.1167/14.4.2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Feng G, Gan Z, Yi HG, Ell SW, Roark CL, Wang S, Wong PCM, & Chandrasekaran B (2021). Neural dynamics underlying the acquisition of distinct auditory category structures. NeuroImage, 118565. 10.1016/j.neuroimage.2021.118565 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ferdinand NK, Becker AMW, Kray J, & Gehring WJ (2016). Feedback processing in children and adolescents: Is there a sensitivity for processing rewarding feedback? Neuropsychologia, 82, 31–38. 10.1016/j.neuropsychologia.2016.01.007 [DOI] [PubMed] [Google Scholar]
  26. Garner WR (1974). The Processing of Information and Structure. Erlbaum. [Google Scholar]
  27. Gathercole SE (1999). Cognitive approaches to the development of short-term memory. Trends in Cognitive Sciences, 3(11), 410–419. 10.1016/s1364-6613(99)01388-1 [DOI] [PubMed] [Google Scholar]
  28. Geffen G, & Sexton MA (1978). The development of auditory strategies of attention. Developmental Psychology, 14(1), 11–17. 10.1037/0012-1649.14.1.11 [DOI] [Google Scholar]
  29. Gesiarz F, Cahill D, & Sharot T (2019). Evidence accumulation is biased by motivation: A computational account. PLoS Computational Biology, 15(6), e1007089. 10.1371/journal.pcbi.1007089 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Gogtay N, Giedd JN, Lusk L, Hayashi KM, Greenstein D, Vaituzis AC, Nugent TF, Herman DH, Clasen LS, Toga AW, Rapoport JL, & Thompson PM (2004). Dynamic mapping of human cortical development during childhood through early adulthood. Proceedings of the National Academy of Sciences of the United States of America, 101(21), 8174–8179. 10.1073/pnas.0402680101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Gopnik A (2020). Childhood as a solution to explore–exploit tensions. Philosophical Transactions of the Royal Society B: Biological Sciences, 375(1803), 20190502. 10.1098/rstb.2019.0502 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Hauser MD (1999). Perseveration, inhibition and the prefrontal cortex: a new look. Current Opinion in Neurobiology, 9(2), 214–222. 10.1016/s0959-4388(99)80030-0 [DOI] [PubMed] [Google Scholar]
  33. Hazan V, & Barrett S (2000). The development of phonemic categorization in children aged 6–12. Journal of Phonetics, 28(4), 377–396. 10.1006/jpho.2000.0121 [DOI] [Google Scholar]
  34. Holt LL, & Lotto AJ (2010). Speech perception as categorization. Attention, Perception, & Psychophysics, 72(5), 1218–1227. 10.3758/app.72.5.1218 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Horowitz-Kraus T, Schmitz R, Hutton JS, & Schumacher J (2017). How to create a successful reader? Milestones in reading development from birth to adolescence. Acta Paediatrica, 106(4), 534–544. 10.1111/apa.13738 [DOI] [PubMed] [Google Scholar]
  36. Huang-Pollock CL, Maddox WT, & Karalunas SL (2011). Development of implicit and explicit category learning. Journal of Experimental Child Psychology, 109(3), 321–335. 10.1016/j.jecp.2011.02.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Idemaru K, & Holt LL (2013). The developmental trajectory of children’s perception and production of English /r/-/l/. The Journal of the Acoustical Society of America, 133(6), 4232–4246. 10.1121/1.4802905 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Isbell E, Fukuda K, Neville HJ, & Vogel EK (2015). Visual working memory continues to develop through adolescence. Frontiers in Psychology, 6, 696. 10.3389/fpsyg.2015.00696 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kazak AE (2018). Editorial: Journal Article Reporting Standards. American Psychologist, 73(1), 1–2. 10.1037/amp0000263 [DOI] [PubMed] [Google Scholar]
  40. Kemler DG, & Smith LB (1978). Is there a developmental trend from integrality to separability in perception? Journal of Experimental Child Psychology, 26(3), 498–507. 10.1016/0022-0965(78)90128-5 [DOI] [Google Scholar]
  41. Kim S, & Lee D (2011). Prefrontal Cortex and Impulsive Decision Making. Biological Psychiatry, 69(12), 1140–1146. 10.1016/j.biopsych.2010.07.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Kirkham NZ, Rea M, Osborne T, White H, & Mareschal D (2019). Do Cues From Multiple Modalities Support Quicker Learning in Primary Schoolchildren? Developmental Psychology, 55(10), 2048–2059. 10.1037/dev0000778 [DOI] [PubMed] [Google Scholar]
  43. Kolk SM, & Rakic P (2021). Development of prefrontal cortex. Neuropsychopharmacology, 47(1), 1–17. 10.1038/s41386-021-01137-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Kruschke JK (1992). ALCOVE: An Exemplar-Based Connectionist Model of Category Learning. Psychological Review, 99(1), 22–44. [DOI] [PubMed] [Google Scholar]
  45. Kuhn MR, & Stahl SA (2003). Fluency: A Review of Developmental and Remedial Practices. Journal of Educational Psychology, 95(1), 3–21. 10.1037/0022-0663.95.1.3 [DOI] [Google Scholar]
  46. Lane DM (1982). Incidental learning and the development of selective attention. Psychological Review, 87(3), 316–319. 10.1037/0033-295x.87.3.316 [DOI] [Google Scholar]
  47. Liquin EG, & Gopnik A (2022). Children are more exploratory and learn more than adults in an approach-avoid task. Cognition, 218, 104940. 10.1016/j.cognition.2021.104940 [DOI] [PubMed] [Google Scholar]
  48. Love BC, Medin DL, & Gureckis TM (2004). SUSTAIN: a network model of category learning. Psychological Review, 111, 309–332. 10.1037/0033-295x.111.2.309 [DOI] [PubMed] [Google Scholar]
  49. Luck SJ, & Vogel EK (1997). The capacity of visual working memory for features and conjunctions. Nature, 390(6657), 279–281. 10.1038/36846 [DOI] [PubMed] [Google Scholar]
  50. Maddox WT, Chandrasekaran B, Smayda K, & Yi H-G (2013). Dual systems of speech category learning across the lifespan. Psychology and Aging, 28(4), 1042–1056. 10.1037/a0034969 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Manning C, Wagenmakers E-J, Norcia AM, Scerif G, & Boehm U (2021). Perceptual Decision-Making in Children: Age-Related Differences and EEG Correlates. Computational Brain & Behavior, 4(1), 53–69. 10.1007/s42113-020-00087-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Martinez JE, Mack ML, Bauer JR, Roe MA, & Church JA (2018). Perceptual biases during cued task switching relate to decision process differences between children and adults. Journal of Experimental Psychology: Human Perception and Performance, 44(10), 1603–1618. 10.1037/xhp0000552 [DOI] [PubMed] [Google Scholar]
  53. McGeoch JA, & Irion AL (1952). The psychology of human learning (2nd ed.). Longmans, Green, & Co. [Google Scholar]
  54. Minda JP, Desroches AS, & Church BA (2008). Learning rule-described and non-rule-described categories: a comparison of children and adults. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34(6), 1518–1533. 10.1037/a0013355 [DOI] [PubMed] [Google Scholar]
  55. Mulder MJ, Maanen L. van, & Forstmann BU (2014). Perceptual decision neurosciences – A model-based review. Neuroscience, 277, 872–884. 10.1016/j.neuroscience.2014.07.031 [DOI] [PubMed] [Google Scholar]
  56. Munakata Y, Morton JB, & Stedron JM (2003). The role of prefrontal cortex in perseveration: Developmental and computational explorations. In Quinlan P (Ed.), Connectionist Models of Development. Psychology Press. [Google Scholar]
  57. Nahum M, Nelken I, & Ahissar M (2010). Stimulus uncertainty and perceptual learning: Similar principles govern auditory and visual learning. Vision Research, 50(4), 391–401. 10.1016/j.visres.2009.09.004 [DOI] [PubMed] [Google Scholar]
  58. National-Reading-Panel. (2000). Teaching children to read: Evidence-Based Reading Policy in the United States: How Scientific Research Informs Instructional Practices. 2005(1), 209–250. 10.1353/pep.2005.0009 [DOI] [Google Scholar]
  59. Nippold MA (2006). Encyclopedia of Language & Linguistics (Second Edition). Language Acquisition: Article Titles: L, 368–373. 10.1016/b0-08-044854-2/00852-x [DOI] [Google Scholar]
  60. Nittrouer S (2004). The role of temporal and dynamic signal components in the perception of syllable-final stop voicing by children and adults. Journal of the Acoustical Society of America, 115(4), 1777–1790. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Nittrouer S, Manning C, & Meyer C (1993). The perceptal weighting of acoustic cues changes with linguistic experience. Proceedings of the Acoustical Society of America, 94(3), 1865–1865. [Google Scholar]
  62. Nordmeyer AE, Yoon EJ, & Frank MC (2016). Distinguishing processing difficulties in inhibition, implicature, and negation. Proceedings of the 38th Annual Meeting of the Cognitive Science Society. [Google Scholar]
  63. Nosofsky RM (1986). Attention, Similarity, and the Identification-Categorization Relationship. Journal of Experimental Psychology: General, 115(1), 39–57. [DOI] [PubMed] [Google Scholar]
  64. Paulon G, Llanos F, Chandrasekaran B, & Sarkar A (2020). Bayesian Semiparametric Longitudinal Drift-Diffusion Mixed Models for Tone Learning in Adults. Journal of the American Statistical Association, 1–14. 10.1080/01621459.2020.1801448 [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Paulon G, & Sarkar A (2021). lddmm: Longitudinal Drift-Diffusion Mixed Models (LDDMM) (0.1.0) [R package]. https://CRAN.R-project.org/package=lddmm [DOI] [PMC free article] [PubMed]
  66. Pick AD, Christy MD, & Frnkel GW (1972). A developmental study of visual selective attention. Journal of Experimental Child Psychology, 14(2), 165–175. 10.1016/0022-0965(72)90041-0 [DOI] [PubMed] [Google Scholar]
  67. Plebanek DJ, & Sloutsky VM (2017). Costs of Selective Attention: When Children Notice What Adults Miss. Psychological Science, 095679761769300–095679761769300. 10.1177/0956797617693005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Plude DJ, Enns JT, & Brodeur D (1994). The development of selective attention: A lifespan overview. Acta Psychologica, 86(2–3), 227–272. 10.1016/0001-6918(94)90004-3 [DOI] [PubMed] [Google Scholar]
  69. R Core Team (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ [Google Scholar]
  70. Rabi R, Miles SJ, & Minda JP (2015). Learning categories via rules and similarity: Comparing adults and children. Journal of Experimental Child Psychology, 131, 149–169. 10.1016/j.jecp.2014.10.007 [DOI] [PubMed] [Google Scholar]
  71. Rabi R, & Minda JP (2014a). Perceptual Category Learning: Similarity and Differences Between Children and Adults. Proceedings of the 36th Annual Conference of the Cognitive Science Society (CogSci 2014), 2811–2816. [Google Scholar]
  72. Rabi R, & Minda JP (2014b). Rule-Based Category Learning in Children: The Role of Age and Executive Functioning. PLoS ONE, 9(1), e85316. 10.1371/journal.pone.0085316 [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Rasinski TV, & Hoffman JV (2003). Oral reading in the school literacy curriculum. Reading Research Quarterly, 38(4), 510–522. 10.1598/rrq.38.4.5 [DOI] [Google Scholar]
  74. Ratcliff R (1978). A Theory of Memory Retrieval. Psychological Review, 85(2), 59–108. 10.1037/0033-295x.85.2.59 [DOI] [Google Scholar]
  75. Ratcliff R, Love J, Thompson CA, & Opfer JE (2012). Children Are Not Like Older Adults: A Diffusion Model Analysis of Developmental Changes in Speeded Responses. Child Development, 83(1), 367–381. 10.1111/j.1467-8624.2011.01683.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Ratcliff R, Smith PL, Brown SD, & McKoon G (2016). Diffusion Decision Model: Current Issues and History. Trends in Cognitive Sciences, 20(4), 260–281. 10.1016/j.tics.2016.01.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Reetzke R, Maddox WT, & Chandrasekaran B (2016). The role of age and executive function in auditory category learning. Journal of Experimental Child Psychology, 142, 48–65. 10.1016/j.jecp.2015.09.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Richler JJ, & Palmeri TJ (2014). Visual category learning. Wiley Interdisciplinary Reviews: Cognitive Science, 5(1), 75–94. 10.1002/wcs.1268 [DOI] [PubMed] [Google Scholar]
  79. Ritz H, DeGutis J, Frank MJ, Esterman M, & Shenhav A (2020). An evidence accumulation model of motivational and developmental influences over sustained attention. Proceedings of the 42nd Annual Meeting of the Cognitive Science Society. [Google Scholar]
  80. Roark CL, & Holt LL (2019). Auditory information-integration category learning in young children and adults. Journal of Experimental Child Psychology, 188, 104673. 10.1016/j.jecp.2019.104673 [DOI] [PMC free article] [PubMed] [Google Scholar]
  81. Roark CL, Lescht E, Hampton Wray A, & Chandrasekaran B (2022, August 2). Auditory and visual category learning in children and adults. 10.17605/OSF.IO/BMK9V. [DOI] [PMC free article] [PubMed] [Google Scholar]
  82. Roark CL, Paulon G, Sarkar A, & Chandrasekaran B (2021). Comparing perceptual category learning across modalities in the same individuals. Psychonomic Bulletin & Review, 28(3), 898–909. 10.3758/s13423-021-01878-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  83. Roark CL, Smayda KE, & Chandrasekaran B (2021). Auditory and Visual Category Learning in Musicians and Nonmusicians. Journal of Experimental Psychology: General. 10.1037/xge0001088 [DOI] [PMC free article] [PubMed] [Google Scholar]
  84. Robinson CW, & Sloutsky VM (2004). Auditory Dominance and Its Change in the Course of Development. Child Development, 75(5), 1387–1401. 10.1111/j.1467-8624.2004.00747.x [DOI] [PubMed] [Google Scholar]
  85. Rolls ET, Grabenhorst F, & Parris BA (2010). Neural Systems Underlying Decisions about Affective Odors. Journal of Cognitive Neuroscience, 22(5), 1069–1082. 10.1162/jocn.2009.21231 [DOI] [PubMed] [Google Scholar]
  86. Rouder JN, Morey RD, Cowan N, Zwilling CE, Morey CC, & Pratte MS (2008). An assessment of fixed-capacity models of visual working memory. Proceedings of the National Academy of Sciences, 105(16), 5975–5979. 10.1073/pnas.0711295105 [DOI] [PMC free article] [PubMed] [Google Scholar]
  87. Schneider RM, & Frank MC (2016). A speed-accuracy trade-off in children’s processing of scalar implicatures. Proceedings of the 38th Annual Meeting of the Cognitive Science Society. [Google Scholar]
  88. Siu CR, & Murphy KM (2018). The development of human visual cortex and clinical implications. Eye and Brain, 10, 25–36. 10.2147/eb.s130893 [DOI] [PMC free article] [PubMed] [Google Scholar]
  89. Sloutsky VM (2010). From Perceptual Categories to Concepts: What Develops? Cognitive Science, 34(7), 1244–1286. 10.1111/j.1551-6709.2010.01129.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  90. Sloutsky VM, & Napolitano AC (2003). Is a picture worth a thousand words? Preference for auditory modality in young children. Child Development, 74(3), 822–833. 10.1111/1467-8624.00570 [DOI] [PubMed] [Google Scholar]
  91. Smith LB, & Kemler DG (1978). Levels of experienced dimensionality in children and adults. Cognitive Psychology, 10(4), 502–532. 10.1016/0010-0285(78)90009-9 [DOI] [PubMed] [Google Scholar]
  92. Stoet G (2010). PsyToolkit: A software package for programming psychological experiments using Linux. Behavior Research Methods, 42(4), 1096–1104. 10.3758/brm.42.4.1096 [DOI] [PubMed] [Google Scholar]
  93. Stoet G (2017). PsyToolkit. Teaching of Psychology, 44(1), 24–31. 10.1177/0098628316677643 [DOI] [Google Scholar]
  94. Strutt GF, Anderson DR, & Well AD (1975). A developmental study of the effects of irrelevant information on speeded classification. Journal of Experimental Child Psychology, 20(1), 127–135. 10.1016/0022-0965(75)90032-6 [DOI] [Google Scholar]
  95. Visscher KM, Kaplan E, Kahana MJ, & Sekuler R (2007). Auditory Short-Term Memory Behaves Like Visual Short-Term Memory. PLoS Biology, 5(3), e56. 10.1371/journal.pbio.0050056 [DOI] [PMC free article] [PubMed] [Google Scholar]
  96. Vogel EK, Woodman GF, & Luck SJ (2001). Storage of Features, Conjunctions, and Objects in Visual Working Memory. Journal of Experimental Psychology: Human Perception and Performance, 27(1), 92–114. 10.1037/0096-1523.27.1.92 [DOI] [PubMed] [Google Scholar]
  97. Voss A, Rothermund K, & Voss J (2004). Interpreting the parameters of the diffusion model: An empirical validation. Memory & Cognition, 32(7), 1206–1220. 10.3758/bf03196893 [DOI] [PubMed] [Google Scholar]
  98. Werker JF, & Tees RC (1984). Cross-language speech perception: Evidence for perceptual reorganization during the first year of life. Infant Behavior and Development, 7(1), 49–63. 10.1016/s0163-6383(84)80022-3 [DOI] [Google Scholar]
  99. Werner LA (2007). Issues in human auditory development. Journal of Communication Disorders, 40(4), 275–283. 10.1016/j.jcomdis.2007.03.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  100. Wickham H (2016). ggplot2: Elegant Graphics for Data Analysis (3.3.5) [R package].
  101. Yi H-G, Maddox WT, Mumford JA, & Chandrasekaran B (2016). The Role of Corticostriatal Systems in Speech Category Learning. Cerebral Cortex, 26(4), 1409–1420. 10.1093/cercor/bhu236 [DOI] [PMC free article] [PubMed] [Google Scholar]
  102. Zevin JD (2012). A sensitive period for shibboleths: The long tail and changing goals of speech perception over the course of development. Developmental Psychobiology, 54(6), 632–642. 10.1002/dev.20611 [DOI] [PubMed] [Google Scholar]
  103. Zhang Z, & Yuan K-H (2018). Practical Statistical Power Analysis Using Webpower and R (Eds). Granger, IN: ISDSA Press. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Material
NIHMS1873348-supplement-Supplemental_Material.pdf (140.9KB, pdf)

RESOURCES