I have often heard people complain that students don't really get better in aural skills classes. Small improvements are possible, but nobody goes from failing Aural Skills 1 to an A in Aural Skills 4—in a skill-based class, even hard work cannot overcome severe problems.

This idea offends my sensibilities, and yet I wonder if it might be true. Several cognitive capacities, including working memory and executive control of attention, are likely relatively fixed in capacity by the time students enter our classrooms (Melby-Lervåg and Hulme 2013, Posner, Rueda, and Kanske 2007). Both are diminished by stress and anxiety (see, e.g., Liston, McEwen, and Casey 2009, Schoofs, Preuß, and Wolf 2008), so when students drowning in one-credit classes feel unsuccessful, these capacities might decrease over time. It's no wonder that many instructors see a clear stratification of grades between high and low performers that persists or widens as time goes by.

If this is a problem with our students, then there's very little we can do about it. But what if it's a problem with the rigidity of our curricula?

As a student at a liberal arts college, I never took an aural skills class; rather, my aural skills were shaped by diverse experiences: arranging for a cappella groups and local bands, observing and leading ensemble rehearsals, reading music for lessons and ensembles, and of course classes in music theory. Now that I teach aural skills classes, my training and available materials have an almost obsessive focus on two specific tasks: in-class sight singing and dictation. Chapters in Merritt and Castro 2015, for example, give a page of instruction followed by "melodies for performance" and then dictations; Kraft 1999 is dedicated entirely to dictation while Rogers and Ottman 2019 is focused on sight singing; Karpinski 2017a is a sight singing anthology, while the instructor's manual for Karpinski 2017b gives answers only for its dictations. Murphy and McConville's survey of contemporary aural skills teaching found that while other tasks were used, the only two that were reported from virtually every survey participant were sight singing and dictation (2017, p. 209). These tasks are valuable, but they are also limited, and doing dictations several times a week for four semesters is a bit like training to be a soccer player by only going running every day. Athletes must train strength, endurance, speed, strategy, and more, and each athlete ends up with different combinations of advantages. If we musicians are to make the most of our capacities, then our task should similarly be to develop ourselves and our students through diverse activities and tasks, making our musical skills as flexible, quick, efficient, and accurate as possible.

I am convinced that a diversity of experiences, placing skills into different contexts, is important for several reasons. First, I believe that some of our low-scoring students might turn out to have musical strengths that are not revealed by repeating sight singing and dictation over and over. Second, all students would experience those skills in more interesting and diverse ways. This would be good both for low-performing students, who might experience an aural skills curriculum as a challenging exploration of different ways to use their skills instead of an increasingly painful dictation laboratory, and for high-performing students, who would get to apply their skills. Finally, using different contexts should also contribute to students' ability to transfer their skills and knowledge to new settings, a goal advocated elsewhere in this volume (Heinsen and Heinsen 2020).

This essay is intended to start a long conversation about broadening aural skills curricula beyond dictation and sight reading. Such a short format could not possibly examine these tasks and their alternatives with enough rigor to establish a new model of instruction, complete with new sets of in-class and assessment tasks. But in order to facilitate a constructive conversation, I suggest three things: first, a change in focus from tasks (e.g., dictation) to skills (e.g., "attentive hearing"); second, an exploration of possible new activities; and finally, implications for larger curricular design.

The standard task of melodic dictation is limiting as an end in itself, but rich in possibilities if we consider its goals. Karpinski 1990 asserts that melodic dictation is useful because it is the best way of developing "attentive hearing, short-term musical memory, focused concentration on specific passages, and an understanding of the fundamentals of pulse, meter, and tonality, as well as traditional notions of rhythm and pitch discrimination and notation" (221–222), but adds, "were dictation the only activity foisted on aural-skills students, those who show initial difficulties might never improve" (208). If we treat Karpinski's list of skills as our primary goal rather than dictation itself, they can be a starting point for generating new activities that exercise these skills in new contexts and different combinations, potentially making them more flexible and efficient. Many instructors already use sing-backs to isolate hearing and memory, and notation of well-known melodies to isolate analysis and notation, but judging from textbooks, these are typically ungraded preparatory exercises. Grading them would elevate their importance in the classroom.

I would argue that attention and memory exercises are particularly underdeveloped in our pedagogy. These capacities are so strongly linked that we cannot exercise and assess one without the other, but activities with a short time from stimulus to response are more focused on attention while activities with a delayed response will place higher demands on memory. Attention-focused exercises include N-back singing and other activities inspired by working memory research, while virtually all that we do, including dictation, can include a time delay before students notate or perform their response. Chunking is also an important component of memory, but while we often claim that students will improve this ability in our courses and it is to a certain extent inherent in the concepts named in chapter titles of textbooks, we seldom focus on it for its own sake. Karpinski 2017b, for example, promises on p. 44 that the student will learn to hear chunks in "subsequent chapters," but the word "chunk" does not appear again in the text. What if we had units on chunking, asking students to take "chunk dictation" or to improve their ability to quickly divide up a notated melody into easily-remembered units?

We can also focus on skills rather than tasks in sight reading. Even if strategies are practiced in class, evaluating only a few sight-singing hearings spread through a semester encourages both students and instructor to focus on the goal and gives the instructor only a limited view into what might be going wrong. We might instead devote time to a sight-singing workshop where students are asked to focus on the process. I have done this with groups of 8 students, sent to practice rooms with a few pieces of choral literature and instructed not to rehearse any excerpt for more than five minutes; in the last 10 minutes of class, the group discussed what went well and what did not and wrote self or group evaluations that included goals for next time. (I graded these on quality of response to these reflection prompts, again encouraging a focus on process and improvement over perfection.) While this did not replace traditional sight-reading hearings, grading and giving feedback on these evaluations gave me a better window into students' processes and put sight-reading skills into a new and less abstract context, hopefully making them (again) more flexible and efficient.

The suggestions so far are similar to what already goes on in many classrooms, but we could go much further in diversifying what we do. We could ask students to work on their internal auditory imagery, trying to develop the sounds students can hear internally to be more vivid and flexible. We could ask students to work on all of their skills in a rehearsal-like environment, practicing error detection, sight reading, how to give constructive feedback based on the nature of an error, etc. We could include units on intonation. We could take inspiration from other musical fields such as audio recording (identifying timbres, spatial location of a recorded sound, filters/effects, etc.), conducting (evaluating and correcting balance and blend), music education (asking students to teach music to each other), and more. And of course, while improvisation has been growing in prominence both in scholarship on pedagogy and in recent textbooks, we could further develop our approaches to this important task.

Assessment of these various skills and activities does pose new challenges, but many can fit into traditional aural skills models. (And, as Kleppinger 2017 points out, even traditional testing and grading policies for dictation may not be as objective or valid as we might think.) Individual hearings are common in aural skills classes, and certain tasks will fit best in these, including sing-backs, N-back singing, and testing of intonation. Chunk dictation can work similarly to traditional dictation, and so long as "allowable" types of chunks have been well-defined for students (say, "for this dictation, you should indicate scalar passages, arpeggios, and skips, and for each, the boundary solfege syllables"), grading may even be more objective than for melodic dictation. Many other tasks can be shaped to fit either the individual hearing model or the written response model: for example, students could write down the kinds of filters or effects they hear applied to a passage of recorded audio, and internal auditory imagery could be tested in a hearing where the instructor sings a pitch, waits a certain amount of time (or, to test manipulation, asks the student to go, say, up two half steps), then asks the student to sing the pitch.

Activities based on situations musicians will confront in real life, such as rehearsals, must be more holistically graded, and I have found self-reflection to be a particularly useful way to guide students to improvement without a high degree of stress. I mentioned above the reflection prompts I use to assess student engagement and improvement in my sight-singing hearings; while grading these may seem rather subjective, the instructor might legitimately choose to grade largely on completion (treating engagement with the activity and reflection as the primary goal) or, more objectively, set grading criteria (say, "makes at least one goal for the following class"; "Shows meaningful engagement with and assessment of the results of last self-evaluation's goal"). Similar self reflections, journal entries, or group reflections can be fashioned for other integrative activities such as leading rehearsals.

To be clear, I believe sight singing and dictation should still be a regular part of aural skills instruction—but ideally alongside more flexible exercises that encourage musical growth, and this will mean less time spent on the traditional two activities. A runner who decided to train for soccer would presumably interleave their running with other elements such as strength training, learning strategy, and practicing with a team; some of these additional activities might even make the person a better runner. Similarly, our classes could mix dictation and sight reading with other activities—and some of these might make our students better at dictation and sight reading. But they will also assess, and develop, the different strengths and abilities our students bring to class, and place all the skills into a variety of contexts.

There is a range of approaches one might take to integrating such diverse activities into an aural skills class, depending on the level of institutional/departmental rigidity, the amount of time and resources the instructor has to consider course alterations, and instructor interest. For example, I made fairly ambitious changes to a recent high-level aural skills course I taught, as our study of modulation involved the following evaluated components, each item representing between 10–25% of the grade for the unit: six recorded melodies sung from notation, an improvisation hearing, a sight-singing hearing, computer-graded quizzes identifying modulations (first yes/no, then by type), a transcription, a short set of dictations, and a small-group rehearsal and recording of a modulating excerpt. Other units, on subjects such as listening for form and performing mixed meter, were similarly diverse. In another recent class, I took a simpler approach by retaining the traditional focus and activities for most of the semester, but interspersing two extended group projects where students could try out their skills in a more career-relevant situation.

I wish this article ended with a fleshed-out model for exactly how this shift in thinking ought to play out, but there is still too much to be decided first. Should diversifying the tasks and skills we focus on change the number of aural skills courses we require? Should there be more, since we are asking students to do more different things, or less, since many of these skills (say, ensemble work) could be integrated into other coursework? Should there be different kinds of aural skills courses? Does changing the tasks we do and assess also require changing the content of these courses? The answers to these questions will likely be different for every instructor and at every institution.

Regardless, if we are unsettled by the gap in scores between our highest- and lowest-performing students, we can put the blame in one of two places. We could use low scores as evidence that those students simply aren't fit for careers in music. Or we could take these instead as inspiration to broaden our curricula, moving past doing our traditional two tasks over and over in order to give our students a true diversity of musical experience.