Step 1: Read what the formula means.
The value here is the excess kurtosis, written as $\dfrac{m_4}{\sigma^4}-3$. The part $\dfrac{m_4}{\sigma^4}$ is the plain kurtosis, and we take away $3$ to compare the curve with a normal one.
Step 2: Fix the normal curve as the bench mark.
For a normal distribution the plain kurtosis is exactly $3$, so its excess kurtosis is
\[ 3-3=0 \]
This is the reason we subtract $3$. The normal curve sits at zero.
Step 3: Decide when the value is positive.
A positive value needs $\dfrac{m_4}{\sigma^4}>3$, which means a sharper peak and fatter tails than the normal curve. Such a curve is called leptokurtic.
Step 4: Check the other names.
A mesokurtic curve and the normal curve both give zero. A platykurtic curve is flatter, so it gives a negative value. None of these are positive.
Step 5: Pick the answer.
Only the sharp peaked case is positive.
\[ \boxed{\text{Leptokurtic distribution}} \]