Interdisciplinarity on the Bench Top

Transcription

1 Interdisciplinarity on the Bench Top Model-based reasoning in bio-science and -engineering university research laboratories Nancy J. Nersessian Georgia Institute of Technology Research Supported by Grants from; NSF REC , REC & SBE , NEH, Radcliffe Institute

2 Research Contributors to Laboratory Studies Project Co-PI: Wendy Newstetter Post Docs: Elke Kurz-Milcke, Lisa Osbeck, Kareen Malone, Yanlong Sun, Barbara Fasse Grad Students: Jim Davies, Etienne Peleprat, Idris His, Chris Patton, Susan Wyche, Ellie Harmon, Arvind Venkataramani Undergrad Students: Matt Labbe, Victoria Smith, Chris Patton, Dillon Mahmoudi, Ellie Harmon, Kristen Baker, Emily Etheridge

3 Why interdisciplinary research laboratories? Innovation communities where cognitive and cultural practices from different disciplines interact Significant sites of learning in science and engineering fields Numerous social and cultural studies of research labs have shown them fruitful locales for thinking about the culture of science Scant research on cognitive practices Hypothesis: rich locales for investigating the interplay of cognition and culture in creative scientific practices

4 Mixed-Method Approach Ethnography & qualitative data analysis ethnographic observations and interviews provide traces of situated activities, tools, and interpretive practices that ground research activity in lab routines, social organization, construction and manipulation of artifacts Cognitive-historical analysis enables following historical trajectories relating problems, tools, models, researchers and traces of how practices originate, develop, change

5 Data Archive Three labs: BME: tissue engineering, neuro-engineering; CS: bio-robotics For each lab: 2 years intensive, 2 years follow-up 800+ hours observing researchers actively working, in research meetings, in journal clubs 170+ interviews Guided tours Audio-visual recordings of group meetings Notebooks, formal and informal research power points, posters, grant proposals, dissertation proposals, dissertations, publications, pictures of artifacts, output of artifacts, sketches/diagrams, pictures of changing lab configurations, s, web sites

6 Bio-Engineering Research Laboratories Lab A: tissue engineering Big problem: develop artificial blood vessels for implantation in the human cardio-vascular system Intermediate: producing constructs that mimic properties of natural vessels, creating endothelial cells from stem cells Lab D: neuro-engineering Big problem: understand the ways neurons learn in brain, create aids for disabilities or to make humans smarter Intermediate: culturing, quieting, and recording neuron arrays, creating feedback environments in which dish of neurons can learn and evolve

7 ONTOLOGY OF ARTIFACTS DEVICES INSTRUMENTS EQUIPMENT flow loop confocal pipette bioreactor flow cytometer flask bi-axial strain mechanical tester water bath construct coulter counter refrigerator beauty and beast sterile hood LM 5 (program) camera computer devices: selectively mimic the human biological environment, engineered sites of in vitro simulation instruments: extract and processes information, generate measured output, enable simulation equipment: assists with manual or mental labor

8 putting a thought into the bench top and seeing whether it works or not In vivo/ex vivo/in vitro framework Devices as models Structural, functional, or behavioral analog of physical objects, processes, situations, events Design incorporates engineering and biological constraints Model-systems sites of experimentation where biological materials and engineered devices interact when everything comes together I would call it a model system [.] the integrated nature, the biological aspect coming together with an engineering aspect, so it s a multifaceted modeling system [ i-i-i21] Problem solving takes place within a fabric of interlocking models distributed in space and time

9 A proposed experiment with a vascular construct model system

10 Interlocking Models in the Vascular Construct Model- System Model of in vivo phenomena as intersection of Cell biology Fluid dynamics Engineered in vitro physical models of selective aspects Devices Model-systems Quantitative models Mental models in vivo and in vitro/ex vivo phenomena device qua in vitro/ex vivo model device qua device quantitative models

11 Model-based Reasoning A process of reasoning through constructing and manipulating models of the same kind with respect to salient dimensions of target phenomena A semantic process in which models are interpretations of target phenomena constructed to satisfy constraints drawn from target domain drawn from one or more modeling source domains include: spatial, temporal, topological, causal, categorical, logical, mathematical Inferences can be specific or generic

12 Kinds of Reasoning Processes (not ordered) Abstraction: limiting case, idealization, generalization, generic modeling Simulation: inferring outcomes or new states via model manipulation Evaluation: goodness of fit (structure, behaviors), explanatory power, implications (empirical, mathematical) Adaptation: constraint satisfaction, coherence, enhanced target understanding, how to model-revise the design

13 Model-Systems are Multi-referential Simulation intent and function points toward biological understanding and systems Make-up and composition points toward engineered systems and structures Participation in BME research points toward experimental in vitro/ex vivo situations Participation in socio-cultural practices points toward a historically situated epistemic community and a learning environment