Psychonomic Bulletin & Review 2007, 14 (3), 512-516 Time (also) flies from left to right Julio Santiago, Juan lupiáñez, elvira pérez and María JeSúS FuneS University of Granada, Granada, Spain Everyday linguistic expressions in many languages suggest that back and front space is projected onto tempo- ral concepts of past and future (as in the sentence we are years ahead of them). The present experiment tested the psychological reality of a different space–time conceptual metaphor—projecting the past to left space and the future to right space—for which there are no linguistic traces in any language. Participants categorized words as referring to the past or to the future. Irrelevant to this task, words appeared either to the left or right of the screen, and responses were given by keypresses of the left or right hand. Judgments were facilitated when word position and response mapping were congruent with the left–past right–future conceptual metaphor. These results are discussed in the context of current claims about the embodiment of meaning and the possible mechanisms by which conceptual metaphors can be acquired. Everyday conversations provide many examples that point to another probably constitutes the experiential basis seem to imply that people locate the past behind them and for our spatialization of time along the back–front axis. the future in front of them, and that they think of time as a There are languages that resort to different ways to proj- movement from one (past) location to another (future) loca- ect space onto time, but again, the back–front axis is the tion (e.g., Back in the 60’s, In the weeks ahead of us; Clark, most productive, since it affords the application of a greater 1973; Lakoff & Johnson, 1980). Although there are alterna- number of image schemas. For example, in Toba (Klein, tive and incompatible ways of conceptualizing time (such 1987) and Aymara (Núñez & Sweetser, 2006), the past is as time is money, Lakoff & Johnson, 1980; Levine, 1997), in front of speakers because it can be “seen,” whereas the the spatial metaphor time is movement along a path seems future is behind because one cannot know (“see”) clearly to be prevalent in many languages. Most of these languages what will happen. The up–down axis is used to a limited ex- use the horizontal back–front axis to map onto temporal tent in Western languages (as in the weeks coming up; Rad- “movement” and conceptualize the experiencer as moving den, 2004) and extensively in Chinese (Boroditsky, 2001; forward from the past into the future: an “ego-moving” per- Yu, 1998). In contrast to the back–front and up–down axes, spective (see Malotki, 1983, for Hopi; Alverson, 1994, for in his cross-linguistic review of spatial metaphors of time, Mandarin, Hindi and Sesotho; Özçaliskan, 2003, for Turk- Radden (2004) found no evidence whatsoever for any use of ish; see Radden, 2004, for a review of data from Western, the left–right axis in lexical items, syntactic constructions, Chinese, West African, and Indian languages). Alternatively, or any other kind of conventional linguistic marker of time: the experiencer may be seen as static, with future events “[The left–right axis] does not seem to offer any sensible frontally approaching and passing him or her by toward the spatial basis for our understanding of time at all” (p. 227). back (a “time-moving” perspective; Clark, 1973). Núñez Although languages often talk of the “approaching month,” and Sweetser (2006) proposed a more elaborate taxonomy, no language uses expressions like “the rightward month.” emphasizing the importance of the reference point. In their Apparently, the only exception to this claim are signed lan- classification, ego-moving and time-moving perspectives guages, which conventionally refer to the left–right axis to are subtypes of the ego-reference-point metaphors, in which place sequences of events (Emmorey, 2001). the deictic origin is placed on the here and now of the In the psycholinguistic literature, there is already direct speaker. There are also time-reference-point metaphors, in evidence available for the back–front time metaphor, both in which the deictic origin is not part of the sequence of events, its ego-moving and time-moving perspectives (Boroditsky, as in Christmas comes after Thanksgiving. 2000; Boroditsky & Ramscar, 2002; McGlone & Harding, Within the conceptual metaphor view (Lakoff & John- 1998), but there are virtually no studies testing the left–right son, 1980, 1999), researchers have suggested that abstract mapping of time onto space. There are, however, some hints concepts (such as time) import their structure through con- that such metaphors should be actually represented in se- ceptual projection from more concrete domains (such as mantic memory, at least in the minds of literate people from space). The building blocks of all concepts are basic image left-to-right writing cultures. In these cultures, not only text schemas that arise from universal perceptuomotor experi- runs from left to right. There is also a highly conventional ences (Johnson, 1987; Mandler, 1992). The image schema association of time to the horizontal axis in graphs, increas- acquired from forward movement of the body from one ing from left to right; comic strips also flow from left to J. Santiago, [email protected] Copyright 2007 Psychonomic Society, Inc. 512 left–Right time flow 513 right, and, in an interesting parallel of the contrast between ceptual metaphors of time and to do it precisely by looking the ego-moving and time-moving perspectives, the pages for the metaphor that maps the past onto left space and the of books move from right to left as we advance in the story. future onto right space, and of which there are no known Tversky, Kugelmass, and Winter (1991) asked American, linguistic traces. We used a simple and straightforward test Arab, and Hebrew children to place three stickers on a piece of this idea. People were asked to judge whether words re- of paper to represent spatial (e.g., three peg dolls arranged ferred to the past or future. Completely irrelevant to this task, in front of them), temporal (e.g., breakfast, lunch, and din- two left–right spatial dimensions were manipulated: Words ner times), preference (e.g., favorite food, a somewhat-liked could be presented to the left or right of the fixation point, food, and totally disliked food), and quantity concepts (e.g., a and past and future responses could be mapped to the left spoonful, a bucketful, and a whole dump truck full of sand). and right hand, respectively, or have the contrary mapping. Only temporal concepts showed a clear linearity effect, with We chose to manipulate both spatial dimensions in order to American children ordering temporal concepts from left to increase our chances of finding the predicted interaction: right and Arab children using a right to left order. Past words should be judged faster when presented to the Placing stickers on paper is a highly off-line and reflex- left and/or responded to with the left hand, and future words ive task, but so far it is the only published experimental should be judged faster when presented to the right and/or study on this issue. Less direct evidence is provided by the responded to with the right hand. If spatial dimensions affect work of Chatterjee and others (Chatterjee, Maher, Gon- categorization behavior in this situation, they are likely to zalez-Rothi & Heilman, 1995; Chatterjee, Southwood, do so by means of the automatic activation of spatial codes & Basilico, 1999; Maher, Chatterjee, Gonzalez-Rothi, & when thinking about past and future. At this point of our re- Heilman, 1995) who studied the flow of events, intimately search, we will not discuss whether the relevant left-to-right connected to the flow of time. These authors showed that metaphor uses ego as its reference point or not, since both English speakers have a left-to-right bias when thinking of possibilities will lead to the same prediction. the agent and patient roles of a sentence: The agent tends to be imagined to the left of the patient. Moreover, actions MeThod like “push” or “pull” are also imagined flowing from left to right. By comparing native speakers of Arabic and Ital- Participants Thirty-two University of Granada undergraduates participated in ian, Maas and Russo (2003) showed that this left-to-right exchange for course credit. All of them were native Spanish speakers bias is associated to writing direction. with normal or corrected-to-normal vision. Another line of indirect evidence is provided by num- ber. It is well established that numerical sequences are Materials represented along a spatial mental line running from left A total of 48 Spanish words were used, half of them referring to to right (Dehaene, Bossini, & Giraux, 1993). People are past time and the other half referring to future time (see the Appen- dix). Seventeen out of the 24 words per condition were tensed verbs faster to respond to small numbers with the left hand than (e.g., dijo, “he said”). Verb lemmas were the same across both the the right hand, and to large numbers with the right hand past and future conditions. They varied in person and number as well than the left hand (the space number association of re- as in tense. Care was exercised to ensure that past and future tense sponse codes [SNARC] effect). A similar effect has been could not be discerned by means of any other characteristic of the shown with ordinal sequences, such as months of the year, verb: Person and number varied randomly, and orthographic marks, letters in the alphabet (Gevers, Reynvoet, & Fias, 2003), such as the accent sign, occurred equally often on verb inflections across categories. The remaining words were temporal adverbs (e.g., and days of the week (Gevers, Reynvoet, & Fias, 2004). mañana, “tomorrow”) and the words pasado (“past”) and futuro The directionality of the number line seems to be linked (“future”). Past and future words were matched in word frequency, to writing direction, since it was also found that the effect length in letters and syllables, and stress placement. faded and even reversed with users of right-to-left lan- guages (Dehaene et al., 1993; Zebian, 2005). Apparatus Finally, Núñez and Sweetser (2006) gave an example of a The experiment was written in MicroExperimental (MEL) code (Schneider, 1988) and run on an IBM PS/2 30 286 personal com- speaker of Aymara gesturing from left to right simultaneously puter. Latencies were measured with millisecond precision. with an Aymara sentence literally meaning day goes, and day after day. The authors suggested that this type of left-to-right Procedure gesturing is common in speakers of many languages when Participants were tested individually in a sound-insulated room. time-reference-point metaphors are used. However, they did The instructions explained that words would be presented one by one not provide statistical analyses of the frequency of occur- on the screen and that their task was to decide whether the word re- ferred to the past or the future. The instructions explicitly mentioned rence of these gestures in the context of relevant expressions, that the words would appear in different positions on the screen, but and their methods were flawed in many other ways (e.g., they that this was irrelevant for the task. There were two response keys, did not blind judges to the hypotheses of the study), factors “f” for the left hand and “j” for the right hand. The experiment con- that make this evidence remain at the anecdotal level. sisted of two halves, which were perfect replications except for the In sum, there is suggestive evidence that literate users of response key to temporal meaning mapping (past to left key and fu- left-to-right written languages map past time onto left space ture to right key was the congruent mapping, whereas the converse was the incongruent mapping). The order of presentation of key map- and future time onto right space. However, the present study ping was counterbalanced over participants. Each half consisted of is the first direct test of the automatic activation of the spa- instructions, 8 practice trials, and two blocks of 48 trials, which were tial left–right axis in the processing of temporal concepts. presented with no break. Materials were divided into two lists that Our aim was to test the psychological reality of spatial con- contained an equal number of past and future words. In the first ex- 514 santiago, lupiáñez, péRez, and funes 1,180 ure 1]. There was a trend toward an interaction between Past meaning and screen position [F(1,31) 5 2.63, MS 5 Future e 1,160 4,459.20, p 5 .11; see also Figure 1]. No other main ef- fect or interaction was reliable in the analysis of latencies. 1,140 The direction of both interactions agreed with our hypoth- eses: Congruent trials (past words presented on the left or y 1,120 nc responded to with the left key, and future words presented e at 1,100 on the right or responded to with the right key) were faster L than incongruent trials by 67 msec with respect to the re- 1,080 sponse position dimension, and by 14 msec with respect to the screen position dimension. Both past and future 1,060 words appeared to be affected by spatial dimensions to a similar extent (see Table 1 and Figure 1). 1,040 Left Right The analysis of accuracy data showed a clear posi- Response tion 3 meaning interaction [F(1,31) 5 11.56, MS 5 e 0.0015, p , .01]. Of less relevance to our hypotheses, 1,140 Past past words were also less accurate than were future words 1,135 Future [F(1,31) 5 7.38, MS 5 0.0052, p , .05]. No other F was e significant at the .05 level. Although the interaction be- 1,130 tween meaning and key was not significant, the pattern of 1,125 means was in the expected direction (responses were 0.5% y c more accurate in the congruent mapping; see Table 1), aten 1,120 showing that there were no speed–accuracy trade-offs. L 1,115 Finally, nowhere did the interaction between screen position and key reach or even approach significance, 1,110 suggesting that there were no Simon-type effects in our data. This result is consistent with the literature that shows 1,105 Simon effects only in the fast section of the RT distribu- 1,100 tion, with slower RTs leading to null or even reversed ef- Left Right Screen Position fects (see, e.g., Mapelli, Rusconi, & Umiltà, 2003). Figure 1. Mean reaction time of correct responses to words diScuSSion with past and future meanings as a function of response position (top panel) and screen position (bottom panel). When participants judged the past or future reference of written words, they were affected by spatial characteristics of the task that were completely irrelevant to their process- perimental block, the words in one list were presented to the left of the fixation point, and words in the other list were presented to the ing goals. The shape of this interaction between space and right of fixation. In the second block, lists switched positions. As- time took the form that was predicted by the left–right con- signment of list to position was also counterbalanced across partici- ceptual metaphor hypothesis. Responses were faster when pants. Therefore, each participant saw the same word four times, once past and future time were mapped onto left and right keys, in each screen position within each key-to-response mapping. respectively, than when the opposite mapping was used. Within each trial, a fixation point was presented centered on the This result is a direct analogy in the domain of time of the screen during 500 msec, followed by a word that remained in view until a response was detected. The fixation point remained in view SNARC effect that was found by Dehaene et al. (1993). during the whole duration of the trial. Words were presented cen- Moreover, words were somewhat faster and clearly more tered on a point located 4 cm to the right or the same distance to the accurate when they were presented on the screen side that left of the fixation point. After a response was detected, there was a was congruent with their temporal meaning: past words blank intertrial interval of 500 msec. on the left, and future words on the right. The fact that both stimulus and response position are design and data Analysis Reaction time (RT) on correct trials and accuracy were recorded irrelevant to the task of judging temporal reference sug- and analyzed as a function of three factors in a fully crossed facto- gests that thinking of the abstract domain of time leads to rial design: meaning (past or future) 3 position (left or right screen position) 3 key (left or right). Latencies below 250 msec and above 2,500 msec were considered outliers and discarded. Table 1 Mean latency (RT, in Milliseconds) and ReSulTS Percent errors (%e) per condition Screen Position Overall error rate was 10%. Out of the correct trials, Left Right 2.87% were outliers and were also discarded from the RT Past Future Past Future analyses. Table 1 shows mean results per condition. There Key RT %E RT %E RT %E RT %E was a clear cross-over interaction between meaning and Left 1,101 11.1 1,159 10.3 1,095 12.9 1,153 9.6 key [F(1,31) 5 9.14, MS 5 31,928.16, p , .01; see Fig- Right 1,144 9.9 1,093 9.3 1,176 12 1,071 6.8 e left–Right time flow 515 the activation of concrete spatial concepts, as predicted by When the present study is considered in the context of the conceptual metaphor view (Lakoff & Johnson, 1980). the literature on conceptual metaphoric representation, a Moreover, present data support the idea that there exists primary question that is left unanswered concerns the ex- a specific mapping between past time and left space and istence of alternative and often incompatible mappings for future time and right space. the same conceptual domains in the same minds. How can An alternative account raised by a reviewer of this a back–front and a left–right mapping of time onto space paper (Daniel Casasanto) suggests that participants may coexist in the minds of Westerners? What are the factors that have noted that the stimuli constitute ordered pairs (e.g., I lead to the activation of one mapping or another? Do con- spoke–I will speak). Gevers et al. (2003, 2004) found that flicts arise and how are they resolved? These questions are ordinal sequences like months, letters, and days of the week thought by many scholars to be key considerations for the tend to be spatialized from left to right. Thus, it might be conceptual metaphor view (Rakova, 2002). Some studies the ordinal relation between the members of each pair and are already approaching this issue by examining the con- not their temporal reference that generates the left–right ditions that lead to the activation of alternative mappings spatialization in our data. Although it may be impossible (Boroditsky, 2000; Boroditsky & Ramscar, 2002). In a rep- to discriminate the correct explanation in the context of the lication and extension of the present results, Torralbo, San- present experiment, some aspects of the data suggest that tiago, & Lupiáñez (2006) have shown that attention plays an a purely ordinal representation of experimental word pairs important role in selecting the relevant space–time mapping. cannot be the whole story. This account predicts that the They manipulated the attention that was payed to either the congruency effect should be stronger in the second half of left–right or back–front spatial frames of reference through the experiment, because the repeated experience with the automatic means (changing task requirements) and found stimulus pairs is supposed to induce and strenghthen their that the attended frame is used to guide the mapping of space representation as ordered pairs. Contrary to such expecta- onto time. Attention to the back–front frame leads to the past tions, follow-up analyses failed to find any sign that the ef- being projected onto back space and the future being pro- fects of either key or screen position congruency increased jected onto front space, and blocks the projection of past toward the second half of the experiment. This result makes and future onto left and right space. The converse occurs the suggested alternative hypothesis less attractive. when the left–right frame is focused. Recently, Santiago, Present results are inconclusive with respect to several Ouellet, and Román (2007) obtained similar results when issues. First, they do not allow us to assert whether the con- attention was manipulated through voluntary-endogenous ceptual metaphor being used is an ego-reference-point or means (task instructions). Thus, attentional mechanisms a time-reference-point metaphor, in Núñez and Sweetser’s may lie at the heart of the solution to this question. Torralbo (2006) terms. Future research will address this question, as et al. (2006; see also Santiago et al., 2007) outline a theo- well as explore other aspects of the inferential structure of retical framework that is able to account for prior results, the metaphor, such as whether there is a continuous relation draw a number of novel predictions, and provide supportive between space and time along the whole mental “time line” evidence. instead of just the dichotomous left–right relation pres- ently reported. The present data do not allow us to assert AuThoR noTe why the mapping takes this form and not, say, past–right The authors are indebted to Antonio Román, Nieves Rodríguez, and and future–left; however, we suggest (for the reasons dis- Ouafa Bouachra for running the experiment and being such enthusiastic cussed in the introduction) that the cause lies in repeated students. We would like to thank Arthur Glenberg and Daniel Casasanto exposure to written language and graphic conventions in for their comments on a prior draft of this article. Address correspon- dence to J. Santiago, Dept. de Psicología, Universidad de Granada, Cam- left-to-right written languages. A direct test of this hypoth- pus de Cartuja, Granada, Spain (e-mail: [email protected]). esis will have to await future cross-linguistic research. If supported by the evidence, future theorizing on conceptual ReFeRenceS metaphor acquisition will need to consider one more route Alverson, H. (1994). Semantics and experience: Universal metaphors to the establishment of cross-domain mappings. So far, ex- of time in English, Mandarin, Hindi, and Sesotho. Baltimore: Johns posure to habitual language has been considered the main Hopkins University Press. determinant of the mappings attested in a given culture Boroditsky, L. (2000). Metaphoric structuring: Understanding time (Boroditsky, 2001; Casasanto et al., 2004; Slobin, 1996). through spatial metaphors. Cognition, 75, 1-28. Boroditsky, L. (2001). Does language shape thought? Mandarin and Eng- Exposure to written language and graphic conventions lish speakers’ conceptions of time. Cognitive Psychology, 43, 1-22. should be considered another important determinant. Boroditsky, L., & Ramscar, M. (2002). The roles of body and mind in Present results not only agree with expectations from the abstract thought. Psychological Science, 13, 185-189. conceptual metaphor view, but also broaden the scope of Casasanto, D., Boroditsky, L., Phillips, W., Greene, J., Goswami, this research program beyond its original boundaries.1 The S., Bocanegra-Thiel, S., et al. (2004). How deep are effects of language on thought? Time estimation in speakers of English, Indo- entire cognitive linguistics enterprise is concerned with nesian, Greek, and Spanish. In K. Forbus, D. Gentner, & T. Regier inferring patterns of thought from the analysis of patterns (Eds.), Proceedings of the 26th Annual Meeting of the Cognitive Sci- of language use. By showing that time can be spatialized ence Society (pp. 186-191). Mahwah, NJ: Erlbaum. in ways that have no corresponding reflection in language, Chatterjee, A., Maher, L. M., Gonzalez-Rothi, L., & Heilman, K. M. (1995). Asyntactic thematic role assignment: The use of a our work extends the relevant cognitive linguistics data- temporal–spatial strategy. Brain & Language, 49, 125-139. base from linguistic conventions to the whole spectrum of Chatterjee, A., Southwood, M. H., & Basilico, D. (1999). Verbs, cultural conventions. events and spatial representations. Neuropsychologia, 37, 395-402. 516 santiago, lupiáñez, péRez, and funes Clark, H. H. (1973). Space, time, semantics and the child. In T. E. Núñez, R. E., & Sweetser, E. (2006). With the future behind them: Moore (Ed.), Cognitive development and the acquisition of language. Convergent evidence from Aymara language and gesture in the cross- (pp. 27-63). New York: Academic Press. linguistic comparison of spatial construals of time. Cognitive Science, Dehaene, S., Bossini, S., & Giraux, P. (1993). The mental representa- 30, 401-450. tion of parity and number magnitude. Journal of Experimental Psy- Özçaliskan, S. (2003). Metaphorical motion in crosslinguistic perspec- chology: General, 122, 371-396. tive: A comparison of English and Turkish. Metaphor & Symbol, 18, Emmorey, K. (2001). Space on hand: The exploitation of signing space 189-228. to illustrate abstract thought. In M. Gattis (Ed.), Spatial schemas and Radden, G. (2004). The metaphor time as space across languages. In abstract thought (pp. 147-174). Cambridge, MA: MIT Press. N. Baumgarten et al. (Eds.), Übersetzen, interkulturelle Kommunika- Gevers, W., Reynvoet, B., & Fias, W. (2003). The mental representation tion, Spracherwerb und Sprachvermittlung—Das Leben mit Mehreren of ordinal sequences is spatially organized. Cognition, 87, B87-B95. Sprachen: Festschrift für Juliane House zum 60. Geburtstag (pp. 225- Gevers, W., Reynvoet, B., & Fias, W. (2004). The mental representa- 238). Bochum: Aks-verlag. tion of ordinal sequences is spatially organized: Evidence from days Rakova, M. (2002). The philosophy of embodied realism: A high price of the week. Cortex, 40, 171-172. to pay? Cognitive Linguistics, 13, 215-244. Johnson, M. (1987). The body in the mind. Chicago: University of Chi- Santiago, J., Román, A., & Oullet, M. (2007). Conceptual meta- cago Press. phors and cognitive flexibility: A review and a theory. Manuscript in Klein, H. (1987). Time in Toba. Word, 38, 173-185. preparation. Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: Schneider, W. (1988). Micro Experimental Laboratory: An integrated University of Chicago Press. system for IBM PC compatibles. Behavior Research Methods, Instru- Lakoff, G., & Johnson, M. (1999). Philosophy in the flesh: The em- ments, & Computers, 20, 206-217. bodied mind and its challenge to Western thought. New York: Basic Slobin, D. (1996). From “thought and language” to “thinking for speak- Books. ing.” In J. J. Gumperz, & S. C. Levinson (Eds.), Rethinking linguistic Levine, R. (1997). A geography of time. New York: Basic Books. relativity (pp. 70-96). Cambridge: Cambridge University Press. Maas, A., & Russo, A. (2003). Directional bias in the mental representa- Torralbo, A., Santiago, J., & Lupiáñez, J. (2006). Flexible concep- tion of spatial events: Nature or culture? Psychological Science, 14, tual projection of time onto spatial frames of reference. Cognitive 296-301. Science, 30, 745-757. Maher, L. M., Chatterjee, A., Gonzalez-Rothi, L., & Heilman, K. M. Tversky, B., Kugelmass, S., & Winter, A. (1991). Cross-cultural and (1995). Agrammatic sentence production: The use of a temporal–spatial developmental trends in graphic productions. Cognitive Psychology, strategy. Brain & Language, 49, 105-124. 23, 515-557. Malotki, E. (1983). Hopi time: A linguistic analysis of temporal con- Yu, N. (1998). The contemporary theory of metaphor: A perspective cepts in the Hopi language. Berlin: Mouton. from Chinese. Amsterdam: Benjamins. Mandler, J. M. (1992). How to build a baby: II. Conceptual primitives. Zebian, S. (2005). Linkages between number concepts, spatial think- Psychological Review, 99, 587-604. ing, and directionality of writing: The SNARC effect and the reverse Mapelli, D., Rusconi, E., & Umiltà, C. (2003). The SNARC effect: SNARC effect in English and Arabic monoliterates, biliterates, and il- An instance of the Simon effect? Cognition, 88, B1-B10. literate Arabic speakers. Journal of Cognition & Culture, 5, 165-190. McGlone, M. S., & Harding, J. L. (1998). Back (or forward?) to the future: The role of perspective in temporal language comprehension. noTe Journal of Experimental Psychology: Learning, Memory, & Cogni- tion, 24, 1211-1223. 1. We thank Daniel Casasanto for bringing this point up. APPendix Materials (Translations in Brackets) Past Future pasado (past) futuro (future) antes (before) después (after) recientemente (recently) próximamente (nearly) anteriormente (previously) posteriormente (subsequently) ayer (yesterday) mañana (tomorrow) anteayer (before yesterday) enseguida (next) antiguamente (formerly) inmediatamente (immediately) dijo (he said) dirá (he will say) tuvimos (we had) tendremos (we will have) hablé (I spoke) hablaré (I will speak) pudimos (we could) podremos (we will be able to) buscásteis (you-plural looked for) buscaremos (we will look for) probásteis (you-plural tried) probaréis (you-plural will try) fuiste (you-singular went) irás (you-singular will go) aparecí (I showed up) apareceré (I will show up) vió (he saw) verá (he will see) pensé (I thought) pensaré (I will think) miré (I looked) miraré (I will look) preguntó (he asked) preguntará (he will ask) decidísteis (you-plural decided) decidiréis (you-plural will decide) quisimos (we wanted) querremos (we will want) hiciste (you-singular made) harás (you-singular will make) creíste (you-singular believed) creerás (you-singular will believe) condujimos (we drove) conduciremos (we will drive) (Manuscript received March 30, 2006; revision accepted for publication July 10, 2006.)