Three figures made of blocks go up on the board: figure 1 is a single block, figure 2 is a square, figure 3 a square — each figure grown from the last by wrapping an L-shaped border down one side and along the bottom. The totals run 1, 4, 9. Two questions, always the same: how many blocks in figure 10? and — the one this course cares about — how many blocks does figure gain when it becomes figure ? The totals are old news; the growth is the new mathematics.
How we play
- Study the figures in silence. See the structure, not the total.
- Predict figure 10 from how you see the pattern growing.
- Defend your count of the growth by pointing at the picture.
Three ways to see the growth
- “Count the L: to grow figure 3 into figure 4, wrap a border of blocks. In general the L holds blocks — two sides of length plus the corner.”
- “Difference the totals: 1, 4, 9, 16 grows by 3, 5, 7 — the odd numbers. Figure 10 is 100, and it took a 19-block L to get there from 81.”
- “Compare to the slopes: the derivative of is , and the L holds . The pattern’s growth is almost exactly twice the side — off by the one corner block, and that lone block matters less and less as the figures grow. A staircase’s growth and a curve’s slope are cousins, and the family resemblance sharpens as the steps shrink.”
One variation
I hand you only the growth numbers of a mystery pattern — say each figure gains 5, then 9, then 13, then 17 blocks — and your group must rebuild the totals and name the degree of the formula behind them. Growth that climbs steadily means totals that climb like a square; constant growth means totals on a line. Recovering the function from its rate of change is the reverse gear this course spends a semester installing — The Slope Detective hands your group a full case of it, with graphs instead of blocks.
The general lives inside the specific
Nobody counts figure 10 block by block. The way you see figure 3 — a square that grows by wrapping an L — is already the formula for the growth of figure . Say what you see, and the algebra writes itself; Derivative Rules does the same thing to whole families of functions, starting from patterns exactly like this one.