28

r- -tassroul 1 1

)p icCatlonswop moDown,,.

BottommmuF

Prnr-essin?N

R ecent research in cognitive neuroscience has

yielded a more comprehensive understanding

of brain function. Some of these diagnostictechniques include the event-related potential

(ERP), which depicts brain electrical activity, and mag-

netic resonance imaging (MRI) and positron emission

tomography (PET), which are particularly sensitiveto the delineation of brain areas. These techniques

poiean insightful look at cognitive process-

es that are not readily studied by behavioralmeasures such as surveys or reaction time.

The advances in neuropsychological re-search have greatly added to our understanding

of top-down and bottom-up processing. However,

top-down and bottom-up processing is a neglected is-sue in education even though it is a critical component

in language, attention, object recognition, and problemsolving. Awareness of its role by teacher and student,

though, may improve student learning.

Defining top-down and bottom-up processingBottom-up processing is stimulus-driven processingwhereas top-down processing is driven by our knowledge,

experience, and intentions, which are relatively voluntaryand stimulus independent. For example, during bottom-

up processing of a word, we notice the orthography and

phonology of the word the arrangements and sounds ofletters in words. Our senses convert these stimuli to neu-ral signals that are sent to primary, low-level areas of the

brain. We then integrate the features in high-level areas.During top-down processing of a word, we can use our

lexical knowledge, for instance, to identify the incomingword as we connect the word to our knowledge of otherrelated words and concepts. Areas of the brain thatcoordinate information would be activated first duringthis high-level thinking.

A similar analysis is possible with attention. Whena child is given a cue about where an image will ap-pear during a video game, the child will use top-downprocessing to selectively attend to that particular area.The child later uses bottom-up processing to analyze

details of the image once it appears so he or she canmake a response decision.

In problem solving, excessive top-down process-ing can impede achieving one's goals. "A Ladder

of Thinking for Students" (Lovrich 2004) usedambiguous pictures in a metacognitive lesson sostudents could discover how using their previousknowledge prematurely-before they finished

bottom-up processing of all of the details-couldcause them to misinterpret words or diagrams.Besides its application to cognition, this up-down

directional nature of our thinking also manifestsitself in neurophysiological measures. Intrigu-ing findings from recent MRI, PET, and ERP

Up

Bottom-upprocessing occurs

when we sensebasic features ofstimuli and thenintegrate them.

Top

Top-downprocessing occurswhen previousexperience andexpectationsare first used torecognize stimuli.

Bottom Down

January 2007 29

h Lovrich

studies of attention and language continue to be criticallyevaluated for evidence of top-down and bottom-up pro-

cessing [for example, Federmeier and Kutas (2001), Frith(2001), Lovrich, Cheng, and Velting (2003), and Deaconet al. (2004).] See Figure 1 for an illustration of bottom-up

and top-down processing.

Relevance to educationThe existence of neurophysiological indicators of a cogni-tive process does not make an idea educationally fruitful.

However, there is concern in science education that as-sessments reflect not only a student's scientific knowledgebut other abilities as well (Clerk and Rutherford 2000;

Harlow and Jones 2004). In the classroom, I often havemotivated students who appear to have a deep level of

understanding, yet perform poorly when asked to readquestions and solve problems on an examination. Whenreviewing test corrections, students sometimes miss a

word or part of a figure, suggestive of too much top-

down processing. These observations and recent researchfindings made me suspect that these students were hav-

ing difficulties in the encoding stage of problem solving,when the given features of a problem have to be noticed

and remembered accurately (Davidson and Sternberg1998). This is when bottom-up processing is essential.Helping students become more aware of top-down and

bottom-up processing might enhance their ability to ad-

just their thinking productively during problem solving.Thinking about how one is accomplishing a task,

or metacognitive processing, has usually meant askingchildren to reflect upon their work after they finisha task (e.g., White and Fredericksen 1998). The cur-

rent classroom approach usually involves presenting

and modeling metacognitive techniques for students;however, the context from which the techniques derive

FIGURE I

The directional nature of bottom-upand top-down processing.

1. The diagrams below represent some events in a cell undergoing nor-mal meiotic cell division.

(9A B C

Which diagram most likely represents a new cell resulting from mei-otic cell division of the cell shown above?

C DA

2. If cell 2 has 30 chromosomes, cell 3 should have

A) 30 chromosomesB) 15 chromosomes --- + "H : "C) 45 chromosomesD) 60 chromosomes 1 2 3 4

3. Which diagram best represents part of the process of sperm formationin an organism that has a normal chromosome number of eight?

10

A® CB

4. The diagram to the right can be used toillustrate a process directly involved in

A) tissue repairB) recombinationC) sexual reproductionD) meiosis

D

N

importance is usually not explained.This is analogous to students learninga laboratory technique without know-ing why they are doing the procedure.Telling why improves comprehension.Knowing why may be one of the mostimportant dimensions for the con-struction of a language-based mentalmodel of a situation (Zwaan andRadvansky 1998) as well as encoding(Calin-Jageman and Ratner 2005).

Using top-down and bottom-up processing in the classroomTo this end, I often include a lessonin my classes on the importance oftop-down and bottom-up processing(Lovrich 2004). Recently, I have triedto help students apply these conceptsto reading examination questions,which seemed to be a circumscribedyet meaningful activity. While studentswere investigating metacognition, theywere also gaining practice answeringquestions on topics such as DNA, ecol-ogy, the cell, diffusion, osmosis, celldivision, and enzymes.

Experimentation, metacognitiveinstruction, and reflective assessmentwere used to "challenge students toaccept and share responsibility fortheir own learning" (NRC 1996, p.32). The learning cycle, consisting ofdiscovery followed by a lesson andconcept application, was modifiedto include metacognitive activities(Blank 2000). Both discovery andconcept application stages endedwith reflective assessment about thestrategic thinking involved, not thescientific content of the questions. Inorder to have published data for com-parison, the reflective assessment wasadapted from the Thinker Tools Cur-riculum, a program involving com-puter simulations, scaffolding, andextended experimentation (White andFredericksen 1998). The assessmentincluded questions on the followingcategories: understanding main ideasand experimentation, being inven-tive and systemic, reasoning carefully,communicating well, and teamwork.

Questions with subtle details indiagrams or wording were chosen

30 The Science Teacher

46 "t

FIGURE 3

Sample student strategies for answering biology questions before and after themetacognitive lesson.

•f--. - - Afte' les-o.

Student Group 11. Read the question.2. Look for key words and mostimportant text.3. Read answers.4. Use your knowledge from biology.5. Choose the best answer forthe question.6. Reread the question.

Student Group 21. Look for key terms (i.e., wordsyou know).2. Read the question, then allanswers before picking your answer3. Double-check your answer withthe question.4. Eliminate all answers you knoware not correct.5. Don't change your first answerunless you're sure that it is wrong.

1. When reading the question, use your pencil and follow the question word for word.2. Underline any important words or phrases.3. When reading the question, cover the picture so you're not distracted.4. Also when reading, underline the most important text.5. Reread the question and then look at the picture more in-depth and pick thebest answer.6. If you do not know the answers by using these steps, take an educated guess.

1. Look for key terms (i.e., words you know) and underline them.2. Read the question and cover your answers and try to think of the answer, and onceyou have, look and see if it is one of the choices.3. Take your time reading and answering the questions; pace yourself.4. If you are stuck on a question, move on to the next one and go back to it later.

5. After you have completed the test, go back to all questions and double-check them

to make sure the answer you chose is the best one.6. Eliminate the choices you know are not correct.Don't change your first choice unless you are 100% sure that it's not correct.

7. Make sure you study for the test; don't slack off and think you know it withoutstudying for it.

from previous New York State Living EnvironmentRegents questions and arranged into six unique pack-ets. A packet had three question sets of four questionseach (Figure 2). Other materials included transpar-encies of ambiguous pictures (Lovrich 2004), reflec-tive assessment questions, PowerPoint, a referencetextbook, Biology: A Human Approach (BSCS 2003),and instruction sheets indicating the goals, activities,and pacing of the lessons planned for the five-periodproject.

My two living environment high school classes of48 students participated in the activity. The studentswere primarily freshmen and sophomores except forfour older students. The approximately 24 students ineach class were divided into six groups designated bythe teacher based on maximizing cooperative behav-ior and balancing academic ability.

During the Discovery phase of a double periodof 80 minutes, each student group examined its firstquestion set of four Regents questions in a uniquepacket. The groups developed a plan-presentedvia PowerPoint-describing how one should answerthe first question set. The goal was not necessarilyfinding the correct answer to a question but rather

developing a strategy about how to answer a question.

Using a neighboring lab group, students tested their

plan with the second question set from their packet.

Finally, the neighboring lab group evaluated the plan

with rating scales, ranging from 1 (not adequate) to 5

(exceptional) with 3 (adequate) as the midpoint, for

each of the reflective assessment categories.

During a 40-minute period the following day,

the students participated in the metacognitive les-

son (Lovrich 2004). During the final double period,

students tried to apply the concepts from the lesson

to the third question set by modifying their original

strategy. Students again tested their new plan on the

same neighboring student group, and the group eval-

uated the new plan. Students then completed their re-

port by graphing the scores on each of the categories

of the assessment questionnaire and answering ques-

tions. Their reports were graded as part of a course

requirement.

Out of 48 participating students, 41 students (85%)handed in their work. The reflective assessment rat-

ing scales were analyzed with Wilcoxon Matched-PairsSigned-Ranks statistical tests. The main finding was

evident in the "being inventive" category. There was a

January 2007 31

FIGURE 4

Sample stu~dent responses to thefollowing question:"How did your strategy or plan change after thelesuonon bo*ttom-up and top-down processing?"

Particip•nt 1: We changed our plan from mostly basic ideasto more inventive ideas.

Participant 2:1 learned to tone down my speed and to takeexceptional time on questions and to get a goodunderstandingof the question.

Participant3: Our plan changed by using techniques that madeyou look at more detail in graphs and other things.

Participan 4: I thought about what they were and then Iunderstood what they were and found out how to use themas a strategy.

Participan 5: Plan first wa very simplified and primitive.Second plan was good and advanced.

significant increase in student perception of the inventive-ness of the strategies (Z=3.07; p=.002), with a score of 4.1after the lesson as compared to a score of 3.6 before thelesson. Two samples of student groups' plans from beforeand after the lesson on bottom-up and top-down process-ing are displayed in Figure 3 (p. 31). At the end of thepacket, students were asked additional questions aboutthe impact of the lesson on their thinking. Some of themost enlightening responses are listed in Figure 4.

Impacting student thinking and learningThe metacognition lesson had its greatest impact on beinginventive, indicating that students thought their neigh-bors' strategies improved in creativity and originality. Inthe White and Fredericksen (1998) study, the "being in-ventive" question was combined with "being systematic"to form a summary design variable. In that study, scoresfor design as well as reasoning carefully and teamworkwere significantly better for the reflective assessmentgroup than the controls. The current demonstration wasnot an experiment with controls and objective assess-ments. Although the present findings are not definitive,they suggest that the metacognitive lesson on bottom-upand top-down processing had a positive impact on stu-dent thinking and learning.

When taking tests, students repeatedly neglect rou-tine words and details in questions. It was a revelationfor some of my weak students that reading slowly anddeliberately-employing bottom-up reasoning-wasa productive strategy. Understanding and reflectingupon processing with material in a specific contentarea has the potential to improve both skill set and

knowledge base simultaneously, facilitating studentimprovement wherever it is needed. Especially in thecontext of a biology course, educational programs thatpromote metacognitive awareness and better under-standing of brain functioning provide rewarding expe-riences for both student and teacher. E

Deborah Lovrich (dlovrich@millerplace.k12.ny.us) is the Director

of Math, Science, Technology, and Business Education at Miller

Place UFSD in Miller Place, New York.

Acknowledgments

This project was funded in part by a Teri Peters Mini-Grant from

the Mid East Suffolk Teacher Center.

References

Biological Sciences Curriculum Studies (BSCS). 2003. Biology: A

human approach. Dubuque, IA: Kendall Hunt.

Blank, L.M. 2000. A metacognitive learning cycle: A better warranty

for student understanding? Science Education 84: 486-506.

Calin-Jageman, R.J., and H. Ratner. 2005. The role of encodingin the self-explanation effect. Cognition and Instruction 23(4):

523-543.

Clerk, D., and M. Rutherford. 2000. Language as a confounding

variable in the diagnosis of misconceptions. International Journal

of Science Education 22: 703-717.

Davidson, J.E., and R.J. Sternberg. 1998. Smart problem solving:

How metacognition helps. In Metacognition in educational theory

and practice, eds. D.J. Hacker, J. Dunlosky, and A.C. Graesser,

47-68. Mahwah, NJ: Erlbaum.

Deacon, D., A. Dynowska, W. Ritter, and J. Grose-Fifer. 2004. Rep-etition and semantic priming of nonwords: Implications for theo-

ries of N400 and word recognition. Psychophysiology 41: 60-74.

Federmeier, K.D., and M. Kutas. 2001. Meaning and modality: In-

fluences of context, semantic memory, organization, and percep-

tual predictability on picture processing. Journal of Experimental

Psychology: Learning, Memory, and Cognition 27: 202-224.

Frith, C. 2001. A framework for studying the neural basis of atten-

tion. Neuropsychologia 39: 1367-1371.

Harlow, A., and A. Jones. 2004. Why students answer TIMMS

science test items the way they do. Research in Science Educa-

tion 34: 221-238.

Lovrich, D. 2004. A ladder of thinking for students: Top-down and

bottom-up processing. The Science Teacher 71(4): 56-59.Lovrich, D., J.C. Cheng, and D. Velting. 2003. ERP correlates of

form and rhyme letter tasks in impaired reading children: A

critical evaluation. Child Neuropsychology 9:159-174.

National Research Council (NRC). 1996. National science education

standards. Washington, DC: National Academy Press.

White, B.Y., and J.R. Fredericksen. 1998. Inquiry, modeling, and

metacognition: Making science accessible to all students. Cogni-

tion and Instruction 16: 3-118.

Zwaan, R.A., and G.A. Radvansky. 1998. Situation models in

language comprehension and memory. Psychological Bulletin

123: 162-185.

32 The Science Teacher

COPYRIGHT INFORMATION

TITLE: Classroom Applications of Top-Down & Bottom-UpProcessing

SOURCE: Sci Teach 74 no1 Ja 2007WN: 0700100742014

The magazine publisher is the copyright holder of this article and itis reproduced with permission. Further reproduction of this article inviolation of the copyright is prohibited. To contact the publisher:http://www.nsta.org/highschool

Copyright 1982-2007 The H.W. Wilson Company. All rights reserved.