Qualities of Speech

The qualities of speech vary with its context. We speak very differently in conversation than we do when we are performing in a play, making stump speeches on the campaign trail, testifying in court, telling a bedtime story, or presenting a unit in the classroom. A good way to examine speech as a communicative medium is to consider it in the context of public speaking, a much-studied form of speech. The two major recommendations you will find in most popular guides to speechmaking highlight the medium's chief strength-its versatility and expressive power-and its chief limitation-the demand it makes on the audience's memory.

Advantages of speech. Perhaps the most basic rule of public speaking is never read your speech. The arguments mostly boil down to a single point: Reading text makes a presentation less natural-sounding, and therefore, less effective.

Natural speech works better because of its enormous expressive power. It offers a wonderfully rich vocal orchestration that enables us to express meanings clearly and energetically, beyond what words and syntax alone can convey. With our voices, we can vary intonation, pace, volume, and pitch to emphasize significant points, clarify intent and point of view, sharpen impact, provide background and emotional tone (such as sarcasm or irony), and even alter meaning. When we speak directly instead of reading text, we also tend to emphasize or clarify our words through physical cues, such as facial expression, gestures, motion, and posture. Speech is also interactive in that we can alter our tone, expressions, and gestures based on our audience's responses.

Limitations of speech. The second basic rule for speechmaking is keep it brief. In Say It in Six (Hoff & Maguire, 1996), the authors argue somewhat hyperbolically that you should never use more than six minutes to say anything. Other oratory experts concur that short and concise is best, but allow that a major speech could run as long as 18 to 20 minutes.

This call for brevity relates to speech's major limitation: its transience. Speech requires our listeners to remember, and it is easy to overload a listener's memory with a long, complex presentation . . . or even a long, complex sentence. Speech's transience and reliance on audience memory is also the reason why expert public speakers use techniques designed to reduce memory load and increase power. For example, Dr. King's "I Have a Dream" speech uses the technique of repetition to great effect. His repeated use of those four famous words provides the structure, rhythmic power, and emphasis that ultimately make the speech more memorable.

Both major characteristics of speech—its rich expressiveness and its transience—are experienced differently by different listeners. In order to find out why and how, we need to explore what understanding speech requires of the brain's recognition, strategic, and affective networks. Of course, speech is a medium used by both teachers and learners in the classroom. For the sake of brevity, we take the standpoint of the learner, focusing on the demands of understanding speech, rather than those of generating it.

Understanding Speech: The Role of Recognition Networks

Spoken communication conveys a sequence of shared-meaning elements that we call words. In the act of understanding speech, recognition networks rely on distributed, parallel processing to interpret many kinds of complex stimuli, while simultaneously processing individual sounds, words, phrases, and intonation. Of course, understanding speech requires both bottom-up processing—interpreting meaning from the flow of sounds—and top-down processing—using prior knowledge and context to predict what words will come next and to make sense of what we are hearing. To clarify how recognition networks process speech, let's first explore the steps required to recognize words and phrases.

Recognizing words requires a complex set of steps including hearing and differentiating individual sounds (also called phonemes), and attaching meaning to words. In speech, phonemes are uttered in rapid sequence. Understanding spoken language requires the ability to hear and distinguish these sounds rapidly to segment the stream of aural input into the words to which we can then attach appropriate meaning.

Barriers to understanding speech include hearing impairments and difficulty segmenting speech-sounds quickly. Research has shown that some children cannot process spoken language quickly enough to segment normal speech into phonemes. It's only when the speech is slowed dramatically (using a computer) that these children can learn to segment speech into the individual sounds that make up words (see Tallal et al., 1996).

Semantic recognition—the ability to attach meaning to individual words—is another critical part of understanding language and another place where barriers may arise. Some of the words we hear cannot be defined based solely on how they sound. The word rock, for example, can mean a type of music, a geological formation, or something we do to soothe a baby. Although most of us can easily assign meaning to the word rock in context ("I stubbed my toe on that rock"), some individuals are unable to connect these words to their meanings—although they can accurately hear and repeat spoken words. This dissociation reflects the fact that recognition networks have two ways of extracting meaning, each one vulnerable to individual differences. They operate in a bottom-up fashion to use phonemic information to recall meaning but when definitions are ambiguous, they must process in a top-down fashion, using contextual information derive meaning (see Federmeier, Segal, Lombrozo, & Kutas, 2000; Ward, Stott, & Parkin, 2000).

As we can see, the task of recognizing spoken words is a complex combination of bottom-upand top-down processing. But recognizing single words is just a tiny fraction of what we must do to interpret speech, which is long strings of words combined into phrases and sentences. We take for granted the understanding of grammar and syntax we apply when we speak and when we listen, and we usually are not aware that we must apply this knowledge to interpret even simple sentences. Difficulties using grammatical constructions can appear independently of other language problems in a disorder called agrammatism (Nadeau, 1988).

Visual cues are also very important for understanding speech because they convey meaning that the bare bones of speech simply cannot express. Facial expressions and gestures provide emphasis, context, and significance to verbal communications. If a student standing outside of the room were to overhear his teacher say to his parents, "Your son is a real trouble-maker," he might become very concerned. However, he probably wouldn't be concerned if he could see his teacher's amused and benign expression—a cue that would help him recognize her joking tone.

Because people who are blind cannot access visual cues, they sometimes miss key facets of spoken conversation. This is also a problem for people with damage to certain parts of recognition networks in the brain's right hemisphere who, although able to see, cannot interpret emotions via facial expressions (see DeKosky et al., 1980). Students with attention deficits may also sometimes fail to interpret visual cues-not because they are incapable of seeing or understanding them, but because they may not properly attend to them (Barkly, 1997; Lyon, 1994).

As this discussion illustrates, understanding speech places extensive demands on a listener's brain. In a classroom context, students must recognize sounds, words, and a variety of aural and visual cues simultaneously in order to attach appropriate meaning to the streams of speech coming from their teacher and their classmates. Thank goodness for parallel processing! But recognition networks are not working alone when students are interpreting spoken language. Listening to and understanding speech also requires us to act strategically and to stay focused and engaged. These are the domains of the strategic and affective networks.

Understanding Speech: The Role of Strategic Networks

It seems obvious that strategic networks, specializing in motor plans and actions, play an essential role in speaking. After all, we have to plan what we're going to say and go through the physical activity of producing speech. However, strategic networks' role in listening is not as immediately clear. Although in a conversation we often must prepare an answer or question while listening to someone speak, the listening itself seems effortless and unplanned. Listening to lectures seems even more passive. How much strategy could listening require?

In fact, listening requires heavy participation of the brain's strategic networks, in part because of the memory demands imposed by speech's transience. To gain meaning we must actively remember what we hear. And despite experts' advice that speechmakers "keep it short," we are often subjected to long speeches, particularly in academic settings! Organizing a long, continuous stream of speech into meaningful segments, placing these segments in context with prior knowledge, and engaging tactics to remember new concepts presented in the speech (taking notes, for example) are all strategic processes that are vital to understanding what we hear. A variety of brain imaging studies corroborate that what we call "active listening" is a significant act of cognition that engages modules throughout strategic networks (Gabrieli, Poldrack, & Desmond, 1998; Posner & Pavese, 1998; Smith, Jonides, Marshuetz, & Koeppe, 1998).

Differences in the ways our strategic networks operate manifest as functional differences in how well each of us can learn from information presented through speech. Students with executive-function disorders, students who have attention deficit hyperactivity disorder (ADHD), and students with other, subtler difficulties rarely have trouble recognizing speech as it is occurring. However, they often have trouble understanding speech because they cannot enact the strategies and skills necessary for active listening, learning, and remembering.

"Verbal working memory" is one strategic function fundamental to speech comprehension. Individuals with impaired working memory functions (for example, some individuals with epilepsy) may be fully capable of recognizing speech but nevertheless struggle to comprehend it because their frontal networks cannot "hold on" to spoken language long enough (Jambaque et al., 1993).

Concentration is another key factor. Understanding speech requires a listener to devote attention selectively and to screen out irrelevant stimuli. Various deficiencies in strategic networks affecting attention can seriously undermine speech comprehension. For example, some patients with Parkinson's disease-a condition characterized by degeneration of the frontal cortex-exhibit an impaired sentence comprehension that appears to be due to a loss of selective attention (Grossman, 1999).

It is interesting to note that students with strong strategic listening skills seem to be able to compensate somewhat for problems with speech recognition. A good memory can help a student retain information long enough to support comparatively slow recognition processes. Further, strong top-down strategic skills help learners predict, hypothesize, and fill in gaps in what they hear. In this way, both strengths and weaknesses in students' strategic networks affect how well they are able to understand speech.

Understanding Speech: The Role of Affective Networks

The demands speech places on memory also call on affective networks. In order to remember what they hear, listeners must stay actively engaged while they listen. Following the advice of the experts, speakers sustain listeners' attention by appealing to their interests, fears, hopes, and senses of humor. Thus, the affective dimension of language—the emotional content that is carried by words, but is distinct from the words themselves-is important for a speaker to consider.

Because affective networks are distributed, different modules process the emotional content conveyed by intonation, facial expression, and gesture. Thus, different people may have trouble with different aspects of the affective processing of speech. For example, neurologist and author Oliver Sacks describes a patient who could understand words perfectly but could not decode the expressive character of speech. When listening to someone speak, she was unable to discern an angry tone from a cheerful or sad one. This meant that unlike most people, she could not rely on affective cues to clarify ambiguous sentences and therefore, had to insist that those around her use unambiguous words. Strict adherence to "proper words in proper places" was the only way she could understand what a speaker intended to say (Sacks, 1985, p. 79).

Damage to speech recognition networks can disrupt the ability to distinguish the speaker's tone of voice (sad, angry, happy) while sparing literal word comprehension (Tucker, Watson, & Heilman, 1977). These kinds of selective deficits further illustrate the distributed, modular nature of language processing. Individuals differ in their ability to employ and interpret these emotional cues, just as they differ in their ability to understand semantic content or to apply strategies for listening and remembering.

Students with emotional difficulties (whether these difficulties have resulted from situational issues or inherent traits) can find listening very challenging. Emotional difficulties alone may impede a student's ability to use information conveyed through speech, in part because the feelings themselves can demand students' attention and make it difficult to concentrate on other matters.

As this discussion has outlined, speech, a seemingly simple medium that most of us take for granted, requires highly complex, rapid processing on the listener's part. With access to nearly a dozen separate channels of communication, including words, gesture, intonation, facial expression, pitch, volume, and pauses, speakers can convey their ideas with great intensity, sharp clarity, and strong emphasis. Because meaning is conveyed through several channels simultaneously, speech supports great subtlety and nuance, much like a complex piece of music. Expression or intonation can contradict words, creating irony or humor. Alternately, expression and intonation can reinforce words, creating emphasis and clarity.

The richness and power of speech make it an excellent medium for communication and teaching. This richness requires extensive processing by the three modular, interconnecting networks in the brain, and learners present myriad subtle strengths and weaknesses as listeners. However, the transience of speech that's inherent in the way we process sound presents a variety of challenges and potential barriers to learners. As teachers, when we are aware of these barriers, we can adjust our teaching methods and materials to support every student's learning. This may involve supplementing speech with other traditional media when conveying concepts or applying the power of digital media to provide additional support.

With the fundamental qualities of speech established, let's move on to text and images, the two other traditional classroom media. Although each offers its own advantages and challenges for students, both depend on the processing of light.