Yes, it is a good Internal Assessment question, particularly for IB Chemistry or Food Science, because it links a controlled variable (flour type) to a measurable dependent variable (internal temperature) and tests a real mechanism: starch gelatinization and protein-driven water binding shift the heat profile of a baking cookie. The question is answerable with a kitchen thermometer, a calibrated oven, and a consistent dough recipe. Its main weakness is that internal temperature alone does not prove chemical differences, so the design needs replication and controlled moisture.
A cookie's internal temperature is not a single fixed number. It rises along a curve shaped by evaporation at the surface, water bound to starch and protein, sugar dissolution, and fat melting. Different flours change how much water is bound and how quickly starch gelatinizes, which shifts the shape of that curve. That makes the keyword question scientifically legitimate rather than a cooking gimmick. The sections below explain the mechanism, the numbers you can expect, the edge cases that ruin reproducibility, common misconceptions, and a concrete experimental protocol you can defend in an IA.
Mechanism: why flour type shifts the temperature curve
Flour is mostly starch (roughly 70 to 75 percent by weight) and protein (about 8 to 14 percent, depending on type), plus small amounts of fiber, lipids, and ash. When dough is heated, water migrates into starch granules, which swell and gelatinize. Gelatinization begins around 140°F to 160°F (60°C to 71°C) for wheat starch in excess water, and the granules absorb water and lose crystallinity. This phase change consumes energy and changes the heat capacity of the matrix, so the rate of temperature rise slows near that window. A flour with more damaged starch or finer particle size hydrates faster and reaches gelatinization earlier in the bake.
Protein matters too. Gluten-forming proteins absorb water and form a viscoelastic network. Higher-protein flours (bread flour at about 12 to 14 percent protein) bind more water per gram than lower-protein flours (cake flour at about 7 to 9 percent). Bound water has lower mobility and lower vapor pressure than free water, so it evaporates less at the surface. Less evaporative cooling at the surface means the interior can heat faster early in the bake, but the same network also insulates and slows conductive heat transfer. These two effects compete, which is why the measured relationship is not simply "more protein equals hotter cookie."
Sugar and fat modify all of this. Sucrose dissolves in the aqueous phase and raises the boiling point of that solution above 212°F (100°C), so a high-sugar dough can hold interior temperatures above the normal water boiling point without drying out. Fat coats starch and delays hydration, which delays gelatinization and can lower the peak internal temperature reached before the cookie sets. The practical consequence: flour type changes the water-binding and gelatinization timing, and those changes appear as measurable differences in the internal temperature curve and the final center temperature at a fixed bake time.
One level deeper, the key variable is water activity and how it partitions between starch, protein, and sugar. A flour that binds more water leaves less free water to evaporate and to dissolve sugar. That raises the effective boiling point of the remaining solution and slows surface cooling. So the internal temperature at any given minute depends on the balance between bound and free water, not on flour type directly. Flour is a proxy for that balance. This is why a careful IA should measure or control dough moisture, not just label the flour.
Quantified context: what temperatures and differences to expect
Cookies are typically baked at 325°F to 375°F (163°C to 190°C), and a finished cookie center commonly reads somewhere between 175°F and 205°F (79°C to 96°C) when pulled from the oven. The exact value depends on dough moisture, sugar content, thickness, and bake time. A thin, high-sugar cookie can exceed 200°F at the center without burning because the sugar solution raises the boiling point. A thick, low-sugar dough may plateau near 190°F as evaporative cooling dominates.
The differences attributable to flour type alone are usually small, often in the range of 2°F to 10°F (1°C to 6°C) at a fixed time point, and sometimes smaller than the thermometer's own error. That is the counterintuitive detail most thin explainers miss: flour type is a real variable, but its effect on internal temperature is often modest compared with the effects of dough moisture, cookie thickness, and oven calibration. A credible IA must therefore report uncertainty and use replication, not a single cookie per flour.
| Flour type | Typical protein | Water absorption tendency | Expected effect on internal temperature curve |
|---|---|---|---|
| Cake flour | 7–9% | Low | Faster hydration, earlier gelatinization, often lower peak center temp |
| All-purpose flour | 10–12% | Medium | Baseline; moderate binding and moderate peak |
| Bread flour | 12–14% | High | More bound water, slower surface cooling, variable peak |
| Whole wheat flour | 13–15% | High (bran competes) | Bran limits gluten, absorbs water, can raise or lower peak unpredictably |
| Almond flour (gluten-free) | ~20% protein, no gluten | Very high fat, low starch | Different matrix; gelatinization minimal, temperature curve dominated by fat |
| Oat flour | 12–15% | High (beta-glucan) | Soluble fiber binds water strongly, can slow the rise |
These are qualitative tendencies, not fixed constants. The table is a hypothesis generator for your IA, not a result. Report your own measured values and compare them against the direction of these tendencies.
A timeline table helps you see where flour effects show up during the bake:
| Time from start | What happens in the dough | Internal temperature signal |
|---|---|---|
| 0–2 min | Fat melts, dough spreads | Center still near room temperature to 100°F |
| 2–5 min | Water migrates, starch begins to hydrate | Rise accelerates through 120°F–150°F |
| 5–8 min | Gelatinization peak, protein sets | Rate of rise slows near 150°F–170°F |
| 8–11 min | Sugar solution boils, surface browns | Center approaches 180°F–200°F |
| 11–14 min | Structure sets, moisture loss continues | Plateau or slow climb; risk of overbake |
| After removal | Carryover cooking | Center may rise 2°F–5°F for 1–2 min |
The plateau in the middle of the bake is the most informative region for your IA. Flour types that bind more water tend to extend or shift that plateau. Measure at fixed intervals (for example, every 60 seconds) rather than only at the end.
Edge cases: what breaks the correlation
Dough moisture is the largest confounder. If you change flour type but keep added water constant by volume, a high-protein flour will produce a stiffer dough and a low-protein flour a wetter one. That changes thickness and surface area, which changes evaporation and internal temperature independent of flour chemistry. Fix added water by mass relative to flour mass, or better, adjust water to a target dough consistency and report both.
Oven calibration and position are next. A 25°F error in oven setpoint is common in home ovens and can shift the internal temperature curve more than flour type does. Use an independent oven thermometer and rotate trays. Cookie thickness, thermometer probe placement, and probe diameter also matter; a thick probe in a thin cookie conducts heat along the metal and biases the reading.
Sugar and fat content must be held constant across treatments. Brown sugar versus white sugar changes moisture and acidity. Butter versus oil changes melting behavior and water content. If any of these vary, you cannot attribute temperature differences to flour. Whole wheat and gluten-free flours are the hardest edge cases because bran and non-starch polysaccharides alter water binding in ways that are not linear with protein content.
Red-flag warnings for your design:
Changing flour type without fixing added water mass makes the comparison invalid. Using a single cookie per flour gives no estimate of variability; use at least three replicates. Measuring only the final temperature loses the curve, which is where the mechanism lives. Ignoring oven calibration can produce a fake "flour effect" larger than the real one. Using a meat thermometer with a thick probe in a thin cookie biases readings by conduction. Comparing recipes that differ in sugar or fat confounds flour with everything else.
Misconceptions to correct before you design the IA
The first misconception is that internal temperature measures flour chemistry directly. It does not. Internal temperature is an emergent property of the whole dough matrix, and flour is one input among several. The correct claim is correlational: flour type is associated with a shift in the temperature curve under controlled conditions.
The second is that higher protein always means a hotter center. Protein binds water, which reduces evaporative cooling and can raise the early curve, but the same network can slow conductive heat transfer. The net effect depends on thickness and bake time. Report direction, not certainty.
The third is that a fixed final internal temperature proves doneness across flours. Doneness is a combination of starch gelatinization, protein setting, and moisture loss. Two flours can reach the same center temperature with different textures. If your IA only measures temperature, say so in the limitations.
The fourth is that gluten-free flours make the comparison cleaner. Almond flour brings high fat, oat flour brings soluble fiber, and both change water activity in ways that dwarf protein effects. They are interesting but harder to interpret.
Practical actions: a defensible IA protocol
Pick two or three flours with clearly different protein contents, such as cake, all-purpose, and bread flour. Keep sugar, fat, egg, leavening, and mixing time identical. Fix added water by mass relative to flour mass, and also report dough consistency. Use a calibrated oven thermometer and bake at a single setpoint, for example 350°F.
Insert a thin digital probe into the geometric center of each cookie and record temperature every 60 seconds until a fixed bake time. Bake at least three cookies per flour, and randomize tray position to spread oven hotspots. Record cookie thickness before and after baking, and weigh dough portions to control mass.
Plot temperature versus time for each flour, then compare the slope in the 150°F to 180°F window and the center temperature at a fixed time. Report means and ranges, not single values. If the ranges overlap, state that the flour effect is within measurement uncertainty. That honest result is still a strong IA because it tests a mechanism and quantifies its limits.
Reconcile thresholds explicitly: gelatinization starts around 140°F to 160°F, sugar solution boiling sits above 212°F, and finished centers commonly land at 175°F to 205°F. These are different reference points, not competing ones. Gelatinization is a starch transition, boiling is a solution property, and the finished range is an outcome. Your IA should cite which threshold it uses and why.
Limitations and what the design cannot show
Internal temperature is a proxy, not a chemical assay. To connect flour type to starch gelatinization or protein setting, you would need differential scanning calorimetry or texture analysis, which are outside most school labs. Your IA can show correlation under controlled conditions, but it cannot isolate the molecular mechanism. Say that plainly.
Sample size and probe error limit precision. A typical instant-read thermometer is accurate to about plus or minus 1°F to 2°F, and cookie-to-cookie variability can exceed that. With three replicates you can estimate a range but not a tight confidence interval. Treat small differences as inconclusive.
Finally, flour is a variable product. Protein content varies by brand, wheat harvest, and milling. Report the brand and lot if possible, and avoid generalizing beyond the flours you tested. A good IA answers "is this a good question" with a qualified yes: good because the mechanism is real and measurable, but valid only if moisture, thickness, and oven conditions are controlled and uncertainty is reported.