A Note on Language

While the CDC recommends using “person-first” language when referring to individuals with various diagnoses or distinguishing traits (e.g., “person with autism,” “individual with a disability”) in an effort to ensure that the individual is not defined by the diagnosis, label, or trait, many persons and groups prefer “identity-first” language, such as “autistic person” or “disabled individual.”15 Similar to many members of the Deaf community, these individuals feel that the features that motivated the diagnosis or labels are inherently tied to their identities.16 Recognizing that identity is deeply personal, both expressions will be utilized interchangeably throughout this work.

In addition, this paper will attempt to address both autism and intellectual disability (ID), understanding that each of these terms describes a spectrum and that they are not mutually inclusive. Autism spectrum disorders (ASDs) can and do exist without accompanying ID, and ID can be completely detached from autistic behaviors or ASD diagnoses. However, both groups have been marginalized within studies of curiosity, and the research methods that augment the studies of one may also bolster studies of the other. And in many of the severest forms of either label, these diagnoses exist together, and, thus, one must account for the superposition of these spectra as well.

Finally, I acknowledge that all of the words and phrases employed here today are subject to the “euphemism treadmill,” whereby words slowly evolve in meaning to become offensive, even if they were originally introduced to replace an offensive word.17 Nevertheless, the intention of this work is accessibility in the broadest sense, and I look forward to improved inclusive lexical semantics in future work.

Mark, Age Ten

Mark is a ten-and-a-half-year-old boy with a diagnosis of ASD. He also has a learning disability and has been referred for concerns regarding attention-deficit/hyperactivity disorder (ADHD), but his parents have not sought a diagnosis for the latter.

At two years of age, Mark had only a few functional spoken words. By age four, his language had dramatically improved to include sentences. He could engage in two-way conversations for a few minutes when motivated; however, his day-to-day use of language was still consistent with the typical early to mid-two-year-old level. At ten and a half, he converses at a similar rate and length to his peers, though his topics tend to focus heavily on things that interest him, such as dinosaurs and dragons.

His parents describe him as “very curious,” “analytical,” and “metacognitive”; he is always trying to ascertain how he thinks and operates. He loves to ask questions, including “why” questions, a hallmark of traditional curiosity questions.18 Mark also enjoys experimenting with material properties and the effects of additives like hot water and soap. For example, he will take toy dinosaurs made of different types of material, including foam, rubber, and hard plastic, into the bathroom and will systematically add different amounts of water (e.g., little or lots) at different temperatures (e.g., hot, warm, or cold) with various additives (e.g., soap or shampoo). These experiments are self-driven and often performed surreptitiously, under the perception that his parents are unaware of his actions.

Although it may seem like Mark would perform admirably in typical studies of curiosity, asking him to play in this way or prompting him with other materials or activities does not necessarily result in similar exploration. In a classic study by Kreitler, Zigler, and Kreitler,19 children were presented with familiar everyday objects, such as “medium-sized toys of a car, an iron, a telephone, and a piano.” The researchers then counted the number of responses the child used to describe the items, as well as the number of “inspective” manipulations (i.e., visually inspecting the toy), “customary” manipulations (e.g., playing the piano, answering the pretend phone), and “exploratory” manipulations (e.g., trying to take the toy apart or determine how it operates). These variables were combined with others to discern different types of curiosity.

Indeed, Mark has rarely shown interest in typical everyday objects. In fact, Mark’s parents reported not having to childproof the house in the way that most parents must because he never approached the knobs on the stove or other items that typically draw children’s attention. Even though Kreitler and colleagues were one of the first groups to acknowledge that curiosity is “neither unitary nor homogenous” and encouraged evaluating a wide range of curiosity types before characterizing a child as “curious or noncurious,” their proposed curiosity measures based on typical development would almost certainly capture Mark’s weaknesses but not his strengths. He would likely be labeled “not curious.”

In what ways could we capture his strengths? How can we characterize his curiosity in a way that is both personalized and rigorous? Like many autistic individuals, Mark exhibits strong specific interests in certain topics or objects that persist longer and with greater intensity than those of nonautistic individuals.20 For example, for the last five years, Mark has been particularly enthralled by dinosaurs and dragons. Mark’s self-stated goal in life is to become the “first hybrid human-dinosaur,” and he independently uses the internet and other resources to learn about anatomy and genetics with the goal of altering his DNA to give himself wings to fly like a dragon.

These devout interests, or “affinities,” can be leveraged to motivate or evoke exploration and inquiry in ways that might otherwise remain hidden,21 as in the case of the bathroom dinosaur science experiments. Affinities also provide a unique opportunity to investigate how different reinforcement mechanisms alter a child’s exploratory behavior. For example, we have developed a research platform called SPRING that enables customization in the activity, the development level of the activity, and the reinforcement provided by the system (Figure 8.1).22 It is similar in style to Banta’s curiosity box, but wholly customizable and programmable.23 The central module, shown as shape sorters in Figure 8.1, can be removed and replaced with different activities to match the developmental skills and interests of the child. Sensors within SPRING passively record user activity, minimizing the effects of examiner presence or intervention while enabling quantitative objective data collection.

By modifying the parameters of SPRING, one can systematically examine a child’s exploration as a function of various reinforcement parameters or levels of ambiguity. For instance, since Mark loves dragons, SPRING could be programmed such that every time Mark inserts a square followed by two circles, his favorite dragon video appears on the device. Inserting other shapes might produce sounds and lights, but no dragons. Thus the significance of dragons as a variable in his exploratory behavior might be examined. By varying the reinforcement, including no reinforcement, the intrinsic and extrinsic motivational factors behind his exploration and problem solving could also be appraised.24 Similarly, prompts could be shown on the SPRING screen, as displayed in Figure 8.1, or the screen could be left blank to examine his exploration and problem solving without assistance. Since SPRING is highly customizable, it and similar personalizable devices can unlock a whole domain of individualized data-driven curiosity studies.

Gabriel, Age Seven

As discussed earlier, Gabriel struggles with motor planning, language, and general learning. While expressive language is expected to be a lifelong struggle, receptive language has also been markedly challenging. Gabriel did not respond in any discernible way to spoken language until shortly before his fifth birthday, and then it was only to the word “car,” referencing the family car, where a few of his favorite toys were kept. Between ages six and seven, he showed considerable growth, with occasional clear recognition of words like “dinner,” “cow” (for a favorite dancing cow toy), and “watch” (as in, “watch a movie”); however, other words, including his name, continue to produce no overt response.

Undoubtedly, attempts to use language to discern the mechanics and motivations of Gabriel’s mind will fall short. Even with alternative means of communication like an augmentative assistive communication (AAC) device, sign language, gestures, picture cards, and other cues, his vocabulary is limited to a few dozen words. And while communication is a major hurdle, it is not the only one. Standardized assessments consistently rate him either in the most severe categories, or such assessments cannot be used because he cannot complete them in the way they were intended. He has been recruited for over a dozen scientific studies, usually ones advertised specifically for autism, but he has not met the inclusion criteria for any of them. His lack of language, cognitive delays, and motor challenges cement his position outside the community of scientific inquiry.

Yet it is not Gabriel’s responsibility to evolve to fit the mold of science. It is our duty as scientists to develop new ways to understand Gabriel. We need a new research framework that captures meaningful data on how he acts and, maybe someday, on how he thinks. We must ask new questions while jointly creating new ways to answer those questions. For example, how does movement of the stimulus vary the perceived complexity of the stimulus? What is the role of agency in exploration? How does personal interest in the topic or object alter curiosity behaviors?

Gabriel explores his environment primarily through sensorimotor actions, including stroking materials to explore textures, spinning items to explore visual dynamics, or tapping objects to surfaces to explore sounds. He finds visually engaging experiences, such as shadows, waterfalls, and dynamic light displays like scrolling LED tickers captivating. Like many autistic individuals, he seems to possess heightened sensory awareness. Some find this hypersensitivity aversive, but Gabriel seeks it out, suggesting a high threshold for optimal sensory input.

But while his sensory seeking is pervasive, it is also specific. A few years ago, a wide balance beam was built for Gabriel that lit up directly in front of his foot as he stepped on it, designed to provide a more enticing environment for him to practice stabilizing and walking. The colored lights were originally programmed to flash three times with each step, and, although Gabriel generally loves lights, he showed little interest. However, when the lights were changed to chase one another down the LED strip, similar to marquee lights, he eagerly approached the beam, willingly practicing challenging skills in order to activate the lights. Such an experience is reminiscent of the curiosity study with the autistic children manipulating the lever, except here we find that not only does the type of stimuli matter (e.g., lights, sounds, and textures) but the dynamics (e.g., how it moves, how it changes) affect the course of action as well. Therefore, within the realm of neurodiverse curiosity studies, and perhaps within all curiosity studies, we must consider what novel features, such as the dynamics of the stimuli, specifically characterize an individual’s curiosity.

Moreover, it is not only what features or variables characterize a person’s curiosity, but also how that curiosity evolves. Recently, Gabriel discovered that a pillow covered in sequins reflects sunlight from a window, covering the walls in golden sparkles of light (Figure 8.2). Twisting the pillow altered the design, intensity, and location of sparkles on the wall. Upon discovery, he spent more than thirty minutes exploring the various patterns of light, his rapt attention far exceeding his usual concentration on a single task or activity. He returned to the pillow and the window for further exploration on subsequent days. Eventually, he began playing with cars and stuffed animals—toys that had never independently held his interest—on top of the pillow (and on top of the pillow only), seemingly motivated to continue exploring the reflection of light in multiple constructs. His curiosity had a distinctive temporal evolution, growing and changing with time.

Even so, developmental assessments would suggest that this exploration is simply a manifestation of his overt developmental age. But is it infantile, or is it artistic? If Gabriel were able to describe his experience in a way that could be understood and shared, we would almost certainly ascribe it to the latter. With language, Gabriel might share the beauty he sees when the reflections cascade across the wall or wonder aloud why sunlight creates the patterns of light while the living room light does not. Without language, it is easy to assume that his actions are merely basal sensory exploration. How could we know differently?

We need new metrics. Berlyne clearly delineates between “perceptual” (stimulus-based) and “epistemic” (knowledge-based) curiosity.25 The former is the major driver in the exploration of novel stimuli, one shared by animals, infants, and adults alike. In that case, however, the physiological arousal induced by the novelty of the stimuli is expected to decrease after repeated exposure, which, in turn, reduces interest in the stimuli.26 Yet Gabriel exhibited no such arousal reduction nor a diminishing interest over time. His interest appeared to grow with time, naturally scaffolding and expanding his exploration to different “sparkle” environments and interactions with other toys, suggesting something distinctive from the classical definitions of both perceptual and epistemic curiosity.

While researchers acknowledge that many variants of curiosity have yet to be fully defined, they still rely on measures like questionnaires and prescribed laboratory studies to test their theories.27 But Gabriel cannot yet explain the motivations for his actions with the sparkle pillow using words, nor can he share his potentially profound sensory experiences in a way that can be cataloged or measured. No clinical study of curiosity would capture, or even replicate, this experience, as it appeared to occur almost through happenstance. How can we test these theories of curiosity for neurodiverse children?

Wearable physiological sensors—for example, watches, glasses, headbands, badges, and clothing that can detect or infer the wearer’s physiological signals without involving wires and constrained environments—may enable a modern approach to tackling these questions and theories. These sensors can track the body’s sympathetic nervous system arousal through electrodermal activity (EDA; formerly galvanic skin response).28 They can also monitor heart rate, heart rate variability, skin temperature, posture, and respiration rate.29 In fact, emerging systems do not even require the use of a wearable device; they can passively survey a person’s physiology through standard cameras, combining light and color magnification with machine-learning analysis techniques.30 With these tools, we can capture and characterize Gabriel’s arousal systematically, and we can do so in a naturalistic environment. We can plot his arousal as a function of activity and behavior and monitor it over time. Is his physiological arousal a function of his curiosity, or is his curiosity a function of his arousal? Does his EDA evolve as his play extends beyond sparkles on the wall to include the reflections on trucks and books that he holds above the pillow? If spikes in arousal initially precede his exploratory behavior, how does this relationship change as his play expands?

Gabriel also has a penchant for music. Although his motor delays inhibit his ability to create sophisticated sounds without assistance, he shows heightened awareness and sensitivity to music over other auditory input. Even as a young toddler, he would choose, through the use of picture cards, to watch video recordings of piano concertos and symphonies, including complex compositions by Rachmaninoff and Dvorak, over animated movie clips or other “age-appropriate” offerings. Years later, he freely and extensively explores the different demo tracks on a digital keyboard, using the programming buttons to modify the tempo, rhythm, and dynamics. But the keyboard is complex. It took years of exposure for him to manually operate the multistep tempo and track buttons, and he still lacks agency to deliberately and accurately select demo tracks. He clearly prefers certain tracks, however, because he will guide a friend or family member over to the keyboard and will continue requesting new music until landing on his favorite song. His lack of agency undeniably impedes his expression of musical curiosity.

Would a control panel with simpler, one-step buttons enable a more extensive, and perhaps more genuine, manifestation of his curiosity? Such a panel would have the added bonus of being able to track button presses and other actions, providing an objective log of his musical exploration over time. Could the differing exploration between the two systems—the complex keyboard and the simpler control panel—measure curiosity as a function of agency? Multiple panels could be built with relative ease, each one increasing slightly in complexity or capabilities by offering more songs or features. Does Gabriel always prefer the simplest access to music, suggesting that it is more about sensory feedback than exploration, or does his interest seem to expand when more options are presented? As with previous studies that examine the optimal levels of complexity, what is the threshold after which exploration declines?31 Similarly, what is the relationship between physical exertion and curiosity? If the buttons are spaced farther apart, are they explored with the same intensity or frequency? Such measures would be valuable not only to other children with special needs but also to any educational or community setting serving a wide range of children and abilities.

Becca, Age 6

Becca is a six-and-a-half-year-old autistic girl, described by her mother as “happy,” a “free spirit,” and the “most confident person I know!” She dives into experiences without restraint and cares little for social etiquette. Becca will unabashedly arrive at school wearing costumes or outfits inspired by her favorite television characters, regularly choosing clothing based on its character-appeal and style over practicality. It is difficult to talk her out of an outfit with long sleeves and a sweater, even on a hot summer day, hinting that she can be equal parts creative and rigid in her actions and expressions.

Becca is a prolific artist, and she has reportedly never paused for even an instant before taking a marker to a piece of paper and beginning to draw. She can fill the pages of entire sketchbooks in rapid succession, never hesitating between one drawing and the next. She has no shortage of ideas, and always seems to know exactly what she wants. And while Becca’s confidence and sense of self are marked assets, she also does not feel motivated to do things that do not interest her, which can make academic work and other obligations difficult. For example, Becca will often spend recess exploring leaves and trees and making tiny piles of wood chips. Asking her to leave recess to do an activity of someone else’s choosing can prove challenging. Likewise, assessments and studies that require her to complete predetermined activities, including curiosity studies that examine her “spontaneous exploration” of a standardized selection of toys, are likely to be met with resistance and may elicit a performance not indicative of her true capabilities.32

Nevertheless, she exhibits undeniable exploratory behavior. Although she is not verbally curious like Mark, Becca will roam around her grandparents’ large yard, climbing rocks, looking at leaves, and bringing grass to her cheek to experience its texture and scent. She experiences the world in a decidedly physical way and can spend hours engaged in what appears to be a very compelling and reinforcing internal world. Some days she will sit and talk to herself (not always in words that others might understand), often laughing aloud at something she is thinking about.

This type of curiosity may be difficult to capture in a laboratory setting, and the few studies that have attempted to do so in the natural environments have relied on recorded narrations of behavior,33 which is time consuming to transcribe, quantify, and scale. Alternatively, multimodal data streams from video, audio, and wearable sensors enable rich new sources to objectively analyze naturalistic behavior and play.34 Combining off-the-shelf cameras and open-source data-processing tools, it is possible to track body pose,35 interactions, and gestures,36 as well as facial expressions37 and affect,38 without the use of specialized equipment or clinical settings. If tolerated, wearable eye-tracking glasses can also help monitor gaze and attention,39 differentiating the minute examination of a pile of wood chips from the inspection of ants marching home after a long day’s work. Machine learning and artificial intelligence enables statistical analysis of these records, including extensive datasets over days and weeks, allowing us to build computational models of behavior and curiosity. These models can help form a baseline of curiosity activities and emotions for individuals like Becca. We then can hypothesize how small changes might affect her curiosity in large ways, taking care to join her world instead of pigeonholing her into ours.

Although Becca can and does occasionally speak, her language remains limited, and approximately 90 percent of her conversations are “scripted” from favorite television characters and scenes. While some of this scripting is appropriate to the situation at hand, observers who are unfamiliar with her nuanced references may not understand the context and may interpret her communication as gibberish or “nonfunctional.” Abstract language remains difficult for her, and she does not ask or answer “why” questions. Yet she has entire television episodes memorized.

She is motivated by these television shows and seeks out opportunities to watch them, think about them, or draw pictures of them. As with many autistic affinities, the relationship she has with the shows and their characters can provide insight into her internal thought processes. For example, almost every picture she draws is a scene or a riff off a scene from an episode of a show. Likewise, if she repeats a line from a favorite episode and a parent or friend says the next line, her eyes will light up and she will smile gleefully as if to say, “Yes! You get it!” It is like a secret handshake, building trust and camaraderie, unlocking the passageway to her precious thoughts and feelings. The songs and plots from these shows can also help with transitions between activities, social situations, and difficult tasks, like acknowledging the need to go to the bathroom when in the middle of a favorite game. They seem to help organize her world, like a Rolodex of situations she can call upon to help process and relate to the emotions and experiences around her.

But these shows and characters may be more than just a conduit to her lush inner world. Becca’s nuanced relationship with them suggests that she is conceptualizing information and making connections far beyond what her words or general actions may indicate. She may be exploring whole theoretical worlds in her head that can only be perceived if one joins her world and explores with her.

So how could we build neurodiverse curiosity studies for Becca? How is her curiosity expressed in the kingdom of her characters? If her characters ask “why” or “I wonder” questions, how does that influence the wonder she expresses through any modality—actions, scripting, or spontaneous speech? What if puppets of her characters explore her real world with inquisitive attitudes and actions? Does she join in? Does she imitate them or learn from them? How do small changes to the ways these characters explore affect how Becca incorporates their words or behaviors into her own life? Even with her favorite shows, she tends to relate more strongly to one character than the rest—for example, Prince Wednesday from Daniel Tiger or Gekko from PJ Masks. How does her assimilation of the information change when Prince Wednesday demonstrates curiosity behaviors versus the default protagonist Daniel Tiger? What is the effect of using words or songs versus only actions? These results may generalize to broader curiosity studies, especially in the context of curiosity and technology,40 but until we understand the most fundamental drivers of these exceptional individuals, we will never know.

This chapter proposed a strengths-based research framework for neurodiverse curiosity studies. Through three distinct case studies, we explored how individual manifestations of curiosity may depend on interests, environment, and skills, and we noted how typical approaches to the study of curiosity may fail for autistic and neurodiverse children. We then described how novel methods and engineering solutions could personalize the approach to elicit and measure curiosity and examined how small parametric changes could significantly affect exploratory behavior outcomes. With these examples and techniques, we invite researchers to form and test new hypotheses, probing how curiosity evolves over time, how it is motivated by personal interests, and how it can be expressed both with and without language. We look eagerly to future studies that expand our understanding of curiosity in all individuals.