Tone envelope influences memory Running head: TONE ENVELOPE INFLUENCES MEMORY

Name that percussive tune: Ecological sounds improve cognitive performance

Michael Schutz University of Virginia Jeanine K. Stefanucci, Andrew Carberry, & Amber Roth The College of William & Mary Word Count: 3984 (including author note, footnotes, and text) Abstract: 150 References: 26 Address correspondence to: Michael Schutz Department of Psychology University of Virginia 102 Gilmer Hall, Box 400400 Charlottesville, VA 22904-4400 Phone: 434-243-5534 E-mail: [email protected]

1

Tone envelope influences memory

2

Abstract Tones with flat envelopes (e.g., the sounds produced by a touch-tone phone) are popular in both psychological research and human-computer-interface design. However, here we show that tones with percussive amplitude envelopes (e.g., the sound of two wine glasses clinking) may be better suited to certain cognitive tasks. In Experiment 1, participants were asked to associate various household objects with 4-note melodies constructed of tones using either flat or percussive envelopes. Those hearing percussive tones correctly recalled 62% more sequenceobject associations. In Experiment 2, participants hearing percussive tones required 42% fewer learning blocks without any detriment to the number of associations retained. In Experiment 3, cell phones using percussive tones were estimated to be worth 9% more (a significant difference). We conclude that amplitude envelope plays an important role in learning and memory, a finding with relevance to psychological research on audition, as well as practical relevance for improving human-computer-interface design.

Key words: tone envelope, amplitude envelope, auditory perception, audition, memory, timbre, paired associates

Tone envelope influences memory

3

Name that percussive tune: Ecological sounds improve cognitive performance A sound’s amplitude envelope describes its intensity over time. Humans are highly attuned to differences in envelope, as they are useful in a variety of perceptual tasks including segmenting speech and recognizing phonemes (Moore, 1997). Sounds produced by percussive events (Figure 1a) decay exponentially, indicating that the energy dissipated naturally due to friction and/or air resistance. Percussive sounds are heard in various forms whenever objects collide as the natural consequence of an impact event. In contrast, the indefinite sustain of tones with flat envelopes (Figure 1b) requires a continual driving force. Although our natural environment does contain continually driven sounds (e.g. howling wind, growling animals, sustained musical instruments) these natural sounds are never static.

INSERT FIGURE 1 HERE Tone Envelope Affects Perceptual Organization The acoustic differences between tones with different envelopes but identical spectral (frequency) information are relatively small; however they produce strikingly different perceptual experiences. For example, playing a piano tone backwards using a digital recorder reverses its envelope while preserving its spectral information. The result of this change is a categorical shift in timbre, producing a note that sounds more like a pipe organ than a piano (Houtsma, Rossing & Wagenaars, 1987). Recently, other research has shown that tone envelope can affect performance on certain perceptual tasks. Schutz and Lipscomb (2007) have demonstrated that expert percussionists employ an audio-visual illusion, using changes in physical gesture length (i.e. the “up-down-up” motion

Tone envelope influences memory

4

used to strike the instrument) to alter audience perception of musical note duration. For example, a given note sounds longer when presented with a physically longer gesture than when paired with a physically shorter gesture. This result is surprising given that it contradicts the well established notion that vision does not influence audition in temporal tasks such as judging event duration (Welch & Warren, 1980, Walker & Scott, 1981). Schutz and Kubovy (2008) have shown that the reason for this contradiction stems largely from differences in tone envelope. They paired a dot moving so as to mimic a visual impact with sounds exhibiting two types of envelopes – percussive (e.g., the sound of two wine glasses toasting, similar to the percussive sounds used by Schutz and Lipscomb) or flat (e.g., the artificial sound made when holding down a key on a cellular phone, similar to the sounds generally used in psychological research). They found that visual information influenced judgments of only the percussive, but not the flat tones. This suggests that the widely accepted view regarding auditory dominance in duration judgments (generally tested using only sounds with artificial flat envelopes) may not actually apply to sounds with percussive envelopes, sounds that are surely more representative of those heard during the evolution of the perceptual system. Nairne and Pandeirada (2008) suggest that memory systems may be organized to process survival-relevant stimuli over other stimuli. As percussive sounds often signal events of survival interest (e.g. hearing two objects collide could signify danger), it is possible that sounds with percussive tone envelopes may activate domain-specific processes tuned to associate them with visible objects. If so, sounds with flat envelopes (which are not encountered in nature) would likely fail to capitalize on this specialized process.

Tone envelope influences memory

5

Tone Envelope: Largely Ignored but Potentially Important The clear perceptual differences between tones with flat and percussive envelopes raises some provocative questions given that traditional investigations of sensory integration rely almost exclusively on tones with flat envelopes (Alais & Burr, 2004; Shams, Kamitani & Shimojo, 2002; Shipley, 1964; Walker & Scott, 1981; Welch, Dutton, Hurt & Warren, 1986). However, we believe this may be indicative of larger differences in the perception and cognition of sounds based upon envelope shape. Such differences would have significant theoretical implications, given the widespread use of flat tones in a wide variety of auditory research paradigms ranging from auditory scene analysis (Bregman, 1990) to the statistical learning of tone sequences (Saffran, Johnson, Aslin & Newport, 1999) to the integration of sensory information (Alais & Burr, 2004; Walker & Scott, 1981). Furthermore, a greater understanding of the role of tone envelope also holds the potential for practical applications, given the pervasive use of flat tones in human-computer-interfaces such as cellular phones and personal digital assistants (PDAs). Beyond the categorical differences in processing based on envelope (Houtsma et a., 1987; Schutz & Kubovy, 2008), there is reason to suspect percussive tones may actually lead to superior cognitive performance. First, percussive tones exhibit an exponential decay, allowing the perceptual system to “predict” their offset. Predicting the offset of a flat tone is not possible, as their indefinite and ambiguous sustain offers no information regarding tone duration prior to the tone’s actual offset. Therefore, flat tones likely require more focused attention to process, distracting from the task of learning sequence-object associations. Given that previous research suggests that enhanced attention during encoding can result in differences in memory

Tone envelope influences memory

6

performance (Jenkins & Postman, 1948), it is possible that differences in attention may translate into differences in memory. The discovery that percussive tones are in some way privileged from a cognitive perspective would have both theoretical and practical implications. Theoretically, it is not clear whether other experiments based exclusively on stimuli made from flat tones accurately describe the way natural sounds are processed. Practically speaking, electronic devices frequently use tones with flat envelopes to communicate with users (e.g. the sound of a phone key being pressed, watches beeping, PDAs reminding users of appointments). If other types of tones are better suited for communication, their adoption could greatly enhance the quality of humancomputer interactions. The use of sound in such contexts frequently requires users to remember associations between sounds and target messages. Therefore, we designed a series of experiments to explore whether sequences of percussive tones are more easily associated with target objects than sequences of flat tones, and whether consumers would prefer devices using them. Experiment 1 We used a modified old-new recognition paradigm (Mandler, 1980; Tulving, 1985) combined with a paired-associates paradigm to assess (1) recognition memory for tone sequences, and (2) recall memory for associations between sequences and everyday household objects. Method Participants. Twenty-two (14 female) College of William and Mary undergraduate students participated for course credit.

Tone envelope influences memory

7

Materials. Ten household objects (cell phone, laser distance meter, CD case, digital clock, keys, calculator, remote control, jewelry box, credit card, and camera) were arranged on a table in front of the seated participant. Order of presentation of the objects was randomized, but matched across conditions (i.e.., 10 orderings were chosen at random, then used one time each for both the percussive and flat conditions). Tones. Tones were generated using one of two amplitude envelopes: “percussive” (Figure 1a) and “flat” (Figure 1b). These envelopes were applied to 13 tones arranged chromatically (e.g. adjacent on the piano) from A2 (220 Hz) to A3 (440 Hz) using SuperCollider1, producing 26 unique frequency-envelope combinations (13 percussive and 13 flat tones). Twenty unique, 4-tone melodic sequences were composed. Audacity2 (a free sound editing program) was then used to construct two versions of each sequence: one using percussive tones and one using flat tones. In order to roughly equate the perceptual distance between tones, flat tones were separated by approximately 200 ms3, and percussive tones were organized adjacently. The sequences were then divided into two sets: set A (sequences 1-10) and set B (sequences 11-20). For half of the participants, set A served as the training or “old” sequences and set B served as the “new” sequences. For the other half of the participants the opposite was true. Procedure. Participants were randomly assigned to either the Percussive or Flat condition. Participants sat in front of a table holding the 10 objects. They were given an 1

http://supercollider.sourceforge.net http://audacity.sourceforge.net 3 Although this approach did increase the overall length of flat sequences, it was a necessary concession to ensure the clear perception of tone onset when using tones of matched acoustic length without introducing differences in the perceived spacing between individual tones. 2

Tone envelope influences memory

8

overview of the experiment and were told explicitly to learn the sequence-object associations as they would be tested after the break. Study. The experimenter selected an object and placed it in front of the participant, played the randomly-assigned sequence three times, and then moved the object to the back of the table. Participants were not allowed to rearrange objects, but could pick up/touch the target object while listening to its associated sequence. This procedure was repeated for each object. Break. Participants were told they would be evaluated after a short break during which time they were allowed to play computer games (either minesweeper or solitaire). The experimenter left the room during this break, returning approximately 6 minutes later. The objects were on the table next to the participant during this time. Evaluation. Participants heard 20 sequences in random order: 10 “old” sequences from the study phase and 10 “new” not previously heard. For each sequence, the participant was asked to decide whether the sequence was “old” or “new,” and to indicate confidence on a scale from 1 (“not at all confident”) to 6 (“very confident”). For all sequences identified as old, they were asked to recall the corresponding object and to provide a confidence rating (same 1-6 scale). This procedure was repeated, without feedback, for all 20 sequences and took approximately 10 minutes. Results and Discussion Recognition Memory. Recognition memory was not statistically different across either of two indices of memory:

Tone envelope influences memory

9

1) The number of hits (correctly identifying old sequences as “old”), did not differ between conditions (percussive M = 7.09, SD = 1.04; flat M = 6.82, SD = 1.25), t (20) = .55, prep = .654 (Figure 2). 2) A d’ analysis balancing hits and false alarms (incorrectly identifying new sequences as “old”) as described by Snodgrass and Corwin (1988) indicates that this did not differ between conditions (percussive M = 1.00, SD = 0.81) and flat (flat M = 0.88, SD = 0.80), t (20) = .38, prep = .60. Therefore, we conclude that recognition accuracy was unaffected by envelope type. Recall. As shown in Figure 2, for sequences correctly identified as old, participants hearing percussive sequences correctly recalled a significantly higher percentage of associated objects (M = 52.90, SD = 25.94) than participants hearing flat sequences (M = 32.67, SD = 11.98), t (20) = 2.55, prep = .95, d = 1.09. Even though the percussive sequences themselves were no more memorable, they were better associated with target objects. INSERT FIGURE 2 HERE Confidence Ratings. There were no significant differences between the percussive and flat conditions in the average confidence rating assigned either to correct recognition responses, t (20) = -.37, prep = .60, or to correctly recalled objects, t (20) = -.76, prep = .70. These results indicate superior performance in associating percussive sequences with target objects. This suggests that percussive tones may be better for communicating information in human-computer-interfaces, which often require users to remember associations between sounds and the event/information they are supposed to communicate. These differences cannot 4

For an explanation of the prep statistic, see Killeen (2005).

Tone envelope influences memory 10 be explained in terms of the distinctiveness of the sequences themselves, as both conditions used the same progression of pitches and there were no differences in the recognition of “old” vs. “new” sequences between envelope conditions. Experiment 2 While the first experiment showed a clear recall advantage for sequences of percussive tones, it raises an interesting question. Were the associations between percussive sequences and objects (1) easier to learn, or (2) learned at the same rate but better retained during the break? In order to differentiate, we designed a variation of the first experiment to determine whether sequences of percussive tones were more quickly associated with target objects than sequences of flat tones. This variation added a training phase in which participants were trained on sequence-object associations (with feedback) before the break. They repeated this training period until able to correctly identify 70% of the sequence-object associations. After training, participants had a break and their memories were evaluated in a manner identical to that of Experiment 1. This design allowed us to distinguish between two possible explanations for the superior performance of the percussive tone participants in the first experiment. Namely, whether performance on the final evaluation differed between tone envelope conditions because (1) associations involving percussive tones are easier to learn, or (2) associations based on percussive sequences are better retained over time and independent of ease-of-learning. Method Participants. Thirty (16 female) College of William and Mary undergraduates participated either to fulfill a course requirement or for payment ($5).

Tone envelope influences memory 11 Materials. The stimuli and apparatus were the same as in Experiment 1. Procedure. The procedure for this experiment was similar to that of Experiment 1, starting with a “study phase” in which participants heard all sequence-object associations three times. However, in this experiment learning was evaluated immediately after completion of the first block of study by playing a randomly selected sequence and asking participants to identify the associated object. After responding, we again played the sequence to reinforce the association, providing feedback regarding the correct object when the participant was incorrect. This procedure was repeated for all 10 sequences and their associated objects. Those unable to correctly recall at least 7 of the 10 associations performed another block of training, in which the sequence was played, the associated object was chosen by the participant, and then feedback was given on their response until they could correctly associate 7 of the 10 sequences with the correct object. After meeting this criterion the procedure was identical to that of Experiment 1, including a break after the learning/training phases before the final memory evaluation. The main questions of interest were whether there were differences in the (1) number of learning blocks required to meet the 7/10 criterion, and (2) retention of the associations after a break, as examined during the evaluation phase. Results and Discussion Participants hearing percussive tones required significantly fewer learning blocks (M = 1.67, SD = .98) to reach the 70% criterion than participants hearing flat tones (M = 2.87, SD = 1.60), t (28) = 2.48, prep = .95, d = 0.90 (Figure 3). Yet performance in the final evaluation phase was unaffected. Once again, recognition of “old” vs. “new” sequences did not differ between the conditions, t (28) = - 0.33, prep = .59, nor did the d’ values of those in the percussive (M =2.12, SD = .84) and flat (M =1.81, SD =.81) conditions, t (28) = -1.04, prep=.76. There was also no

Tone envelope influences memory 12 difference in recall of the associations between the percussive (M = 5.33, SD = 2.53) and flat (M = 5.07, SD = 2.19) conditions, prep=.59. Analysis of the confidence ratings revealed that participants in the percussive condition were slightly more confident in their recognition of the sequences, t (28) = -2.09, prep = .92, d = 0.76. Therefore, despite hearing 42% fewer learning trials, participants in the percussive condition were slightly more confident (and no less accurate) in their responses. From these results, we tentatively conclude that the superior performance of participants in the percussive condition in Experiment 1 may have been due to quicker learning, rather than better retention. Importantly, the additional trials given to participants in the flat envelope condition did not lead to any differences in task performance and/or confidence. Therefore, while associations based on percussive sequences may also be better retained, we did not observe any indication of this here (possibly due to ceiling effects in the memory measures). Experiment 3 The potential practical applications for percussive tones suggested from the previous experiments raise the question of whether consumers would also find them more pleasant. Therefore, we allowed participants to compare the tone sequences back-to-back by presenting them as cell phone “ring tones.” Cellular and touch-tone phones always produce tones with flat envelopes to communicate information regarding dialing. Given that familiarity generally reliably predicts judgments of liking (Berryman, 1984; Innes, 1974; Harrison, 1969), we would expect participants would be inclined to rate the flat sequences as more “likeable” when played back-to-back with the percussive tones. Method

Tone envelope influences memory 13 Participants. Twenty (9 female, 11 male) College of William and Mary undergraduates participated for course credit. Materials. The participants were presented with descriptions and pictures of two cell phones that differed only in display and orientation direction (the green display phone opened to the right, the yellow to the left). Procedure. Participants were told that we were interested in how cell phone companies could increase their appeal to customers. They were told they would be asked to indicate which of the two phones they preferred after studying their respective features in succession. Participants were shown the picture and the description of the first phone on a piece of paper (order of presentation was randomized). The experimenter informed participants they would be presented with the sound associated with a missed call in order to demonstrate the phone’s sound quality. Participants then heard either a percussive or flat sequence taken from Experiment 1. When participants indicated they were finished reviewing the features, they were given unlimited time to study the phones, the second cell phone was presented in a similar manner. The pairing of phone with tone sequences was counterbalanced. Participants indicated which of the phones they preferred with a forced-choice decision, and were asked to estimate the amount they would be willing to pay for each if shopping for a new phone. Finally, the experimenter played each of the tone sequences again and asked the participants to rate on a scale from 1 (“absolutely detest”) to 100 (“adore”) how much they liked the tones, as well as which style they preferred (if any). Results and Discussion Overall Preference. Eighty-five percent of the participants preferred the percussive phone to the flat phone.

Tone envelope influences memory 14 INSERT FIGURE 4 HERE Ring Tone Preference. Participants rated the percussive ring tone (M = 55.65, SD = 17.34) as more likeable than the flat ring tone (M = 30.10, SD = 16.26), t (19) = 7.18, prep = .99, d = 1.52 (Figure 4). When asked to indicate whether they preferred one of the tone sequences, 90% of the participants chose the percussive sequences, 10% said they preferred neither, and no participants preferred the flat sequences. Estimated Value. Participants were willing to pay $5.25 more for the percussive phone (M = 63.75, SD = 33.87) compared to the flat phone (M = 58.50, SD = 32.45), which as indicated by a paired-samples t-test was a significant difference, t (19) = 7.18, prep = .97. There was no difference in willingness to pay for the percussive phone based on the order in which participants were presented with the phones, t (18) = .55, prep = .63. Despite a lifetime of associating phones with flat tones, participants overwhelmingly preferred the use of percussive tones. After hearing both types of tone sequences in Experiment 3, 90% of the participants preferred the percussive tones and for 85% this translated to a preference for the actual phone itself. Furthermore, as a result of differences in tone envelope, phones using percussive tones were estimated to be worth 9% more. General Discussion This series of experiments shows that percussive sequences are more easily (requiring 42% fewer learning trials in Experiment 2) and effectively (yielding 62% increase in the number of items correctly recalled in Experiment 1) associated with target objects than flat sequences. Additionally, percussive sequences were rated as more pleasant than flat sequences, enhancing the perceived value of products using them (Experiment 3). Given that both the flat and

Tone envelope influences memory 15 percussive sounds were based on pure tones (and were therefore spectrally identical), the magnitude of these results is quite surprising. The wide-spread use of flat tones in both humancomputer-interfaces and a great deal of psychological research, suggests that these results have implications both practical and theoretical.

Practical Applications These results suggest that using sounds with percussive rather than flat envelopes might significantly enhance the quality of human-computer interfaces. Many devices attempt to use sounds to inform users that a “menu is now open” or that “you’ve got mail.” If our results can be applied to such interface designs, they could lead to shorter learning curves and improved usability. Furthermore, percussive tones appear to enhance perceived quality, leading to more pleasant user experiences. Given our sensitivity to tone envelope and the fact that it has not been previously explored as a useful parameter, there is great potential to use it as a nuanced communicative tool. For example, whereas a certain pitch sequence may identify an incoming call, the shape of each tone’s envelope could also be manipulated to reflect signal strength, allowing a user to decide whether or not to answer the phone based on current circumstances. Given that (1) the acoustic manipulations required to generate percussive tones are trivial and (2) the perceptual consequences for improving both the quality and pleasantness of human-computer interactions are significant, these results should be of interest to a wide community.

Theoretical applications It is possible that the superior performance of participants in the percussive condition can be attributed to the difficulty placed on the perceptual system in processing tones with

Tone envelope influences memory 16 ecologically foreign flat envelopes. If the perceptual system relies on a tone’s rate of decay (in general a useful and informative feature) to process sound, this would pose problems for the perceptual and cognitive processing of tones with flat envelopes, as they offer no such information. Although further research is needed to uncover the precise explanation for the striking differences in performance and likeability between the two types of tones, it is clear from these studies that the traditional reliance on flat tones in a wide variety of psychological research paradigms (Alais & Burr, 2004; Bregman, 1990; Saffran et. al, 1999; Walker & Scott, 1981) should not go unquestioned. Given the differences in learning and memory reported here, further research is need to explore whether experiments relying solely on flat tones accurately describe the way our auditory system functions when processing sounds outside the laboratory.

Final thoughts Although our data illustrates clear differences in the processing of percussive and flat tones, the underlying reasons for these findings are less clear. This is largely a reflection of the lack of research on the role of tone envelope. As shown in Table 1, a search5 of the full content of publications by the Acoustical Society of America reveals that research on tone envelope (274 publications) pales in comparison to that sound pressure (3,618) and frequency (39,427). A parallel search of all publications indexed by PsycInfo produced similar results. Due to this

5

Search results for JASA are from based on the search tool at http://scitation.aip.org/vsearch/servlet/VerityServlet?KEY=ASADL, including the full “bibliographic record” option, with results from PsycInfo based on a similar search. Given that several of these terms are often used in non-auditory research, hits for the latter search were required to additionally include “auditory” in order to be counted. For all three searches, hits for “tone envelope” include all hits using “amplitude envelope” and “temporal envelope.” Physical terms and their perceptual correlates are taken from Rossing, Moore, & Wheeler (2002), page 95.

Tone envelope influences memory 17 relative neglect of the role of temporal envelope in perception and cognition, it is difficult to definitively comment on the underlying reasons for the results we report here. However, it is clear that the effect of temporal envelope on the perception and cognition of auditory information represents a rich and under-researched area of great theoretical and applied interest.

INSERT TABLE 1 HERE

Tone envelope influences memory 18 References Alais, D. & Burr, D. (2004). The ventriloquist effect results from near-optimal bimodal integration. Current Biology, 14, 257-262. Arndt, J., & Reder, L. M. (2003). The effect of distinctive visual information on false recognition. Journal of Memory and Language, 48, 1-15. Berryman, J. C. (1984). Interest and liking: Further sequential effects. Current Psychological Research & Reviews, 3, 39-42. Bregman, A. S. (1990). Auditory Scene Analysis: The Perceptual Organization of Sound. Cambridge: MIT Press. Davis, E. T., Scott, K., Pair, J., Hodges, L. F., Oliverio, J. (1999). Can audio enhance visual perception and performance in a virtual environment? Proceedings of The Human Factors And Ergonomics Society 43rd Annual Meeting: 1197-1201. Ernst, M. O, & Banks, M. S. (2002). Humans integrate visual and haptic information in a statistically optimal fashion, Nature, 415, 429-433. Jenkins, W. O., & Postman, L. (1948). Isolation and spread of effect in serial learning. American Journal of Psychology, 61, 214-221. Harrison, A, A. (1969). Exposure and popularity. Journal of Personality, 37, 359-377. Innes, J. M. (1974). The effect of familiarity and incentive on liking and exploration. European Journal of Social Psychology, 4, 489-494. Killeen, P. R. (2005). An alternative to null-hypothesis significance tests. Psychological Science, 16, 345-353. Houtsma, A.J.M., Rossing, T.D. & Wagenaars, W.M. (1987). Auditory Demonstrations (Philips Compact Disc #1126-061 and text). Melville, N.Y.: Acoustical Society of America

Tone envelope influences memory 19 Mandler (1980). Recognizing: The judgment of previous occurrence. Psychological Review, 87, 252-271. Moore, B. (1997). An Introduction to the Psychology of Hearing (4th ed.). Academic Press: London. Nairne, R., & Pandeirada, J. (2008). Remembering With a Stone-Age Brain. Current Directions in Psychological Science, 17, 239-243. Park, H., Arndt, J. D., & Reder, L. M. (2006). A contextual interference account of distinctiveness effects in recognition. Memory & Cognition, 34, 462-471. Rossing, T. Moore, F., Wheeler, P. (2002). The science of sound (3rd edition). Addison Wesley: San Francisco. Saffran, J.R., Johnson, E. K., Aslin, R. N., & Newport, E. L. (1999). Statistical learning of tone sequences by human infants and adults. Cognition, 70, 27-52. Schutz, M. & Kubovy, M. (2008, July). The effect of tone envelope on sensory integration: support for the 'unity assumption' (A). Journal of the Acoustical Society of America, 123, 3412 Schutz. M., & Lipscomb, S. (2007). Hearing gestures, seeing music: Vision influences perceived tone duration. Perception, 36, 888-897. Shams, L., Kamitani, Y., & Shimojo, S. (2002). Visual illusion induced by sound. Cognitive Brain Research, 14, 147–152. Shipley, T. (1964). Auditory flutter-driving of visual flicker. Science, 145, 1328–1330. Snodgrass, J. G., & Corwin, J. (1988). Pragmatics of measuring recognition memory: Applications to dementia and amnesia. Journal of Experimental Psychology: General, 117, 35-50.

Tone envelope influences memory 20 Tulving, E. (1985). Memory and consciousness, Canadian Psychology, 26, 1-12. Walker, J. T., & Scott, K. J. (1981). Auditory - visual conflicts in the perceived duration of lights, tones, and gaps. Journal of Experimental Psychology: Human Perception and Performance, 7, 1327- 1339. Welch, R. B., DuttonHurt, L. D., & Warren, D. H. (1986). Contributions of audition and vision to temporal rate perception. Perception & Psychophysics, 39, 294–300. Welch, R. B., & Warren, D. H. (1980). Immediate response to intersensory discrepancy. Psychological Bulletin, 88, 638- 667.

Tone envelope influences memory 21 Author Note Michael Schutz, Department of Psychology, University of Virginia; Jeanine K. Stefanucci, Andrew Carberry, and Amber Roth, Department of Psychology, The College of William & Mary. The authors wish to thank Dennis Proffitt for his assistance in designing these experiments, as well as Charissa Beaver and Daniel Paris for their help in collecting the data. Correspondence concerning this article should be addressed to Michael Schutz, Department of Psychology, University of Virginia, 102 Gilmer Hall, Box 400400, Charlottesville, VA 22904-4400, [email protected].

Tone envelope influences memory 22 Figure Captions Figure 1. Percussive and Flat Envelopes. Although both envelopes exhibit a sharp attack, percussive envelopes (left) immediately begin decaying exponentially whereas flat envelopes (right) sustain indefinitely before ending abruptly. Figure 2. Although the number of sequences recognized did not differ based on tone envelope (left panel), the percentage of associated objects correctly recalled for sequences correctly identified as old was 62% greater for those participants listening to sequences of percussive tones (right panel). Error bars represent +/-1 standard error about the mean. Figure 3. Participants hearing sequences using percussive tones required 42% fewer training blocks than those hearing sequences of flat tones. Error bars represent +/-1 standard error about the mean. Figure 4. Phones using percussive tones were rated significantly more likable. Error bars represent +/-1 standard error about the mean.

Tone envelope influences memory 23 Table Captions

Table 1. Relative research focus on frequency, sound pressure, and tone envelope. Results listed separately for (1) these terms alone in Journal of the Acoustical Society of America, (2) these terms along with their strongest perceptual correlates according to Rossing et al. (2002) in JASA, and (3) the terms and their perceptual correlates in all journals indexed by PsycInfo. The first number in each cell indicates the raw number of publications, the second indicates the number of publications as a percentage of all publications in a given column.

Tone envelope influences memory 24

Tone envelope influences memory 25

Tone envelope influences memory 26

Tone envelope influences memory 27

Tone envelope influences memory 28

Search term(s):

JASA Single term

JASA Both terms

PsycInfo Both terms

Frequency (& Pitch)

39,427 (91.0%)

2,298 (89.9%)

3,255 (94.4%)

Sound pressure (& Loudness)

3618 (8.35%)

236 (9.23%)

137 (5.0%)

Amplitude envelope (& timbre)

274 (.63%)

23 (0.9%)

7 (0.6%)