The Water Has More Than One Job

Imagine putting a soft loaf into a plastic bag, closing it, and coming back to bread that resists your fingers. The obvious suspect is missing water. Bread goes hard because it dries out; a bag ought to slow the escape.

It does. That is why the bag is such a useful place to discover that drying is not the whole story.

The question is not whether water matters. It is whether knowing how much water remains tells us what the bread will feel like. Two experiments make that distinction surprisingly concrete: one put ordinary bakery loaves into different storage conditions; the other changed what the crumb could do with its water.

Same wrapping, different aging

In a 2021 study, Thekla Alpers and colleagues bought pan wheat bread and a rye–wheat sourdough from a bakery in Freising, Germany. They compared storage in a bread box, a partly ventilated paper bag, and plastic bags. The particularly revealing comparison used the same plastic wrapping at about 22°C and 8°C—the latter their refrigerator condition. They removed slices daily for ten days and measured both moisture and the force needed to compress the crumb. The slices rested before texture testing to compensate for temperature differences: this was not simply a contest between a cold slice and a warm one. [1, §§2.1–2.4]

The plastic did its expected job. It limited moisture loss better than the other containers. But cooling the bagged bread accelerated its firming relative to the matching room-temperature bags.

For pan wheat bread, the reported ten-day absolute moisture-loss values were similar: 3.5% in room-temperature plastic and 3.1% in refrigerated plastic, with substantial variability. The fitted rate of firming, however, rose from 0.29 to 0.41 newtons per day. The mixed sourdough showed the same direction: similar reported moisture losses, but firming rates of 0.92 and 1.38 newtons per day. [1, Tables 2–3]

Those moisture measurements do not establish identical water loss, and certainly not zero loss. These were bags opened for daily sampling, not perfect sealed-system demonstrations. What they undermine is the simple explanation that the colder bread became firmer because it dried substantially more.

That matters because refrigeration was also doing something useful. No mold was detected in the refrigerated samples by visual inspection or plating during the ten-day experiment, while room-temperature plastic allowed mold growth. Physical aging and microbial spoilage were running on different clocks. Slowing one did not mean slowing the other. That bounded result is not a safety guarantee or a shelf-life rule for another loaf. [1, §3.2]

The bag is not the boundary that matters everywhere

Bread crumb is not just a sponge holding a stock of water. It is an airy, load-bearing structure: swollen starch granules sit within a network that includes gluten and starch released during baking. The flexibility of that network helps determine how readily the crumb yields under pressure. [2, Introduction]

Baking disrupts part of starch's organized structure. During storage, chains of amylopectin—the branched component of starch—can associate again into more ordered arrangements. This process is called retrogradation. It is a rearrangement of molecules, not bread returning to raw flour and not microbial decay. [1, §3.4; 2, §3.3]

Water participates in that rearrangement. It can also move from moist crumb toward drier crust, or change its interactions with components inside the crumb. A bag limits exchange with the room; it does not freeze those internal relationships.

One of water's jobs is plasticization: helping a polymer network remain flexible rather than rigid. Water still present somewhere in a loaf need not be doing the same softening work in the same place. Starch ordering and redistribution of water offer an explanation for why retaining moisture does not stop firming. But these are interacting processes, not a complete molecular diagnosis of each loaf in the storage trial. [2, Introduction, §3.3]

At this point, it would be easy to replace one tidy explanation with another: forget drying; stale bread is simply starch making crystals.

The second experiment is why I would not stop there.

A softer crumb without a smaller starch-aging signal

In 2024, Maria Franco and colleagues studied wheat bread with some flour replaced by psyllium husk, a material with a strong affinity for water. They adjusted the water addition to suit the dough's consistency. This was therefore not a fiber-only experiment with identical water in every recipe: psyllium, hydration and flour replacement changed together. [2, §§2.2–2.3]

The breads were stored in sealed polyethylene bags at 20–22°C. All formulations lost crumb moisture and became harder over time. But formulations with at least 5% of the flour replaced by psyllium remained softer across the storage period. [2, §§2.3, 3.3]

The researchers also used a calorimeter on freeze-dried, milled crumb rehydrated for the test, measuring the heat absorbed when ordered starch structures melted. This thermal signal, attributed to retrograded amylopectin, grew during storage. After seven days, the values across the formulations clustered between 3.0 and 3.2 joules per gram of wheat flour, with no statistically significant differences between them. [2, Table 3]

Softer bread, then, did not require a detected reduction in that particular starch-aging measure.

This does not establish identical starch structures, or make retrogradation irrelevant. It shows why a measurement of starch order is not automatically a measurement of softness. Different crumbs can carry a similar measured thermal signal while responding differently to compression.

The team's nuclear-magnetic-resonance measurements supplied another handle: signals associated with water mobility and its interactions in the crumb. Together with the moisture and texture results, these supported an explanation in which psyllium plus extra hydration maintained more favorable water dynamics and a more flexible network. The precise molecular allocation remains an interpretation, and the changing recipe prevents assigning the effect to psyllium alone. [2, §§2.7, 3.2–3.3]

The detail that changed the question for me was not that one ingredient made bread softer. It was that the softer crumb and the starch-aging signal could part company.

Water retained is not softness retained

The first experiment shows why bulk moisture loss alone cannot explain the observed firming contrast. The second shows why firmness is not interchangeable with one measured consequence of starch ordering. Neither makes water loss unimportant. Neither supplies a single replacement variable that explains everything.

A plastic bag can be doing its job while the bread changes inside it. The water has not merely stayed or gone. It has different places to be, different structures to interact with, and different work to do.

That leaves a more interesting next question than “How do we stop bread drying?” If warming and water addition were independently controlled, how much softness could each recover—and which changes in the crumb would remain? That would test the rearranging structure, rather than just count the water left behind.

Sources and limits

[1] Alpers et al. (2021), “Impact of Storing Condition on Staling and Microbial Spoilage Behavior of Bread and Their Contribution to Prevent Food Waste,” Foods 10, 76. Complete open text. One purchased manufacturing batch and two bread types, with triplicate measurements; not a universal storage or food-safety test.

[2] Franco, Spotti, Gomez and Martinez (2024), “Understanding the influence of the arabinoxylan-rich psyllium (Plantago ovata) husk on dough elasticity and bread staling: Interplay between biopolymer and water dynamics,” Food Hydrocolloids 154, 110099. Published paper, author-institution-hosted PDF, DOI. Two batches of an experimental wheat-bread system; a formulation intervention, not an independent repeat of the refrigerator trial. A non-significant thermal comparison is not proof of equivalence.

I read both main texts, methods, numerical tables and captions. I did not inspect supplementary material or figure images, bake or taste bread, or reanalyze raw data. The opening is imagined. This is an explanation of material aging, not advice about eating stored or moldy bread.